[{"data":1,"prerenderedAt":1635},["ShallowReactive",2],{"active-banner":3,"navbar-featured-partner-blog":24,"navbar-pricing-featured":306,"blog-\u002Fblog\u002Fhandling-100k-consumers-with-one-pulsar-topic":1086,"blog-authors-\u002Fblog\u002Fhandling-100k-consumers-with-one-pulsar-topic":1600,"related-\u002Fblog\u002Fhandling-100k-consumers-with-one-pulsar-topic":1615},{"id":4,"title":5,"date":6,"dismissible":7,"extension":8,"link":9,"link2":10,"linkText":11,"linkText2":12,"meta":13,"stem":21,"variant":22,"__hash__":23},"banners\u002Fbanners\u002Flakestream-ufk-launch.md","StreamNative Introduces Lakestream Architecture and Launches Native Kafka Service","2026-04-07",true,"md","\u002Fblog\u002Ffrom-streams-to-lakestreams","https:\u002F\u002Fconsole.streamnative.cloud\u002Fsignup?from=banner_lakestream-launch","Read Announcement","Sign Up Now",{"body":14},{"type":15,"value":16,"toc":17},"minimark",[],{"title":18,"searchDepth":19,"depth":19,"links":20},"",2,[],"banners\u002Flakestream-ufk-launch","default","zRueBGutATZB0ZnFFHwaEV7F0Di4tnZUHhgOiI4cu6k",{"id":25,"title":26,"authors":27,"body":29,"canonicalUrl":289,"category":290,"createdAt":289,"date":291,"description":292,"extension":8,"featured":7,"image":293,"isDraft":294,"link":289,"meta":295,"navigation":7,"order":296,"path":297,"readingTime":298,"relatedResources":289,"seo":299,"stem":300,"tags":301,"__hash__":305},"blogs\u002Fblog\u002Fstreamnative-recognized-in-the-forrester-wave-streaming-data-platforms-2025.md","StreamNative Recognized as a Contender in The Forrester Wave™: Streaming Data Platforms, Q4 2025",[28],"David Kjerrumgaard",{"type":15,"value":30,"toc":276},[31,39,47,51,67,73,78,81,87,102,109,115,118,124,127,134,140,143,146,157,163,169,172,175,178,184,191,194,197,204,207,210,224,229,233,237,241,245,249,251,268,270],[32,33,35],"h3",{"id":34},"receives-highest-possible-scores-in-both-the-messaging-and-resource-optimization-criteria",[36,37,38],"em",{},"Receives Highest Possible Scores in BOTH the Messaging and Resource Optimization Criteria",[40,41,43],"h2",{"id":42},"introduction",[44,45,46],"strong",{},"Introduction",[48,49,50],"p",{},"Real-time data has become the backbone of modern innovation. As artificial intelligence (AI) and digital services demand instantaneous insights, organizations are realizing that streaming data is no longer optional – it's essential for delivering timely, context-rich experiences. StreamNative's data streaming platform is built precisely for this reality, ensuring data is immediate, reliable, and ready to power critical applications.",[48,52,53,54,63,64],{},"Today, we're excited to announce that Forrester Research has named StreamNative as a Contender in its evaluation, ",[55,56,58],"a",{"href":57},"\u002Freports\u002Frecognized-in-the-forrester-wave-tm-streaming-data-platforms-q4-2025",[36,59,60],{},[44,61,62],{},"The Forrester Wave™: Streaming Data Platforms, Q4 2025",". This report evaluated 15 top streaming data platform providers, and we're proud to share that ",[44,65,66],{},"StreamNative received the highest scores possible—5 out of 5—in both the Messaging and Resource Optimization criteria.",[48,68,69,70],{},"***Forrester's Take: ***",[36,71,72],{},"\"StreamNative is a good fit for enterprises that want an Apache Pulsar implementation that is also compatible with Kafka APIs.\"",[48,74,75],{},[36,76,77],{},"— The Forrester Wave™: Streaming Data Platforms, Q4 2025",[48,79,80],{},"Being recognized in the Forrester Wave is a proud milestone, and for us, it highlights how far StreamNative has come in enabling enterprises to unlock the power of real-time data. In the sections below, we'll dive into what we believe sets StreamNative apart—from our modern architecture and cloud-native design to our open-source foundation and real-time use cases—and how we see these strengths aligning with Forrester's findings.",[40,82,84],{"id":83},"trusted-by-industry-leaders",[44,85,86],{},"Trusted by Industry Leaders",[48,88,89,90,93,94,97,98,101],{},"Companies across industries are already leveraging StreamNative to drive real-time outcomes. Global enterprises like ",[44,91,92],{},"Cisco"," rely on StreamNative to handle massive IoT telemetry, supporting 245 million+ connected devices. Martech leaders such as ",[44,95,96],{},"Iterable"," process billions of events per day with StreamNative for hyper-personalized customer engagement. And in financial services, ",[44,99,100],{},"FICO"," trusts StreamNative to power its real-time fraud detection and analytics pipelines with a secure, scalable streaming backbone.",[48,103,104,105,108],{},"The Forrester report notes that, “",[36,106,107],{},"Customers appreciate the lower infrastructure costs that result from StreamNative’s cost-efficient, Kafka-compatible architecture. Customers note excellent support responsiveness…","”",[40,110,112],{"id":111},"modern-cloud-native-architecture-built-for-scale",[44,113,114],{},"Modern, Cloud-Native Architecture Built for Scale",[48,116,117],{},"From day one, StreamNative was designed with a modern architecture to meet the demanding scale and flexibility requirements of real-time data. Unlike legacy streaming systems that often rely on tightly coupled storage and compute, StreamNative's platform takes a cloud-native approach: it decouples these layers to enable elastic scalability and efficient resource utilization across any environment. The core is powered by Apache Pulsar—a distributed messaging and streaming engine—enhanced with multi-protocol support (including native Apache Kafka API compatibility) to unify diverse data streams under one roof. This means organizations can consolidate siloed messaging systems and handle both high-volume event streams and traditional message queues on a single platform, without sacrificing performance or reliability.",[48,119,120,121,108],{},"Forrester's evaluation described that “",[36,122,123],{},"StreamNative aims to provide a high-performance, multi-protocol streaming data platform: It uses Apache Pulsar with Kafka API compatibility to deliver cost-efficient, real-time applications for enterprises. It appeals to organizations that want a flexible, low-cost streaming solution, due to its focus on scalability and resource optimization, while its investments in Pulsar’s open-source ecosystem and performance optimization make it the primary platform for enterprises wishing to implement Pulsar.",[48,125,126],{},"Our cloud-first, leaderless architecture (with no single broker bottlenecks) and tiered storage model were built to maximize throughput and cost-efficiency for real-time workloads. By separating compute from storage and leveraging distributed object storage, StreamNative can retain huge volumes of event data indefinitely while keeping compute costs in check—effectively providing a flexible, low-cost streaming solution.",[48,128,129,130,133],{},"This modern design not only delivers high performance, but also ensures fault tolerance and geo-distribution out of the box, so enterprises can trust their streaming data is always available and durable. As Forrester’s evaluation noted, StreamNative ",[36,131,132],{},"\"excels at messaging and resource optimization\" and “Its platform supports use cases like real-time analytics and event-driven architectures with robust scalability.","” Our architecture provides the strong foundation that today's real-time applications demand, from ultra-fast data ingestion to seamless scale-out across hybrid and multi-cloud environments.",[40,135,137],{"id":136},"open-source-foundation-and-pulsar-expertise",[44,138,139],{},"Open Source Foundation and Pulsar Expertise",[48,141,142],{},"StreamNative's DNA is rooted in open source innovation. Our founders are the original creators of Apache Pulsar, and we've built our platform with the same open principles: freedom, flexibility, and community-driven innovation. For developers and data teams, this means adopting StreamNative comes with no proprietary lock-in—instead, you get a platform built on open standards and a thriving ecosystem. We offer broad API compatibility (Pulsar, Kafka, JMS, MQTT, and more) so that teams can work with familiar interfaces and integrate StreamNative into existing systems with ease.",[48,144,145],{},"StreamNative is the primary commercial contributor to the Apache Pulsar project and its surrounding ecosystem. We invest heavily in Pulsar's ongoing improvements our investments in Pulsar's open-source ecosystem and performance optimization bolster StreamNative's value. We also foster a vibrant community through initiatives like the Data Streaming Summit and free training resources.",[48,147,148,149,152,153,156],{},"Forrester's assessment noted that StreamNative’s “",[36,150,151],{},"events-driven agents, extensibility, and performance architecture are solid,","” and we're continuing to build on that foundation. ",[44,154,155],{},"We're actively investing in expanding our tooling for observability, governance, schema management, and developer productivity","—areas we recognize as critical for enterprise adoption and where we're committed to accelerating our roadmap.",[48,158,159,160],{},"Being open also means embracing an open ecosystem of technologies. StreamNative actively integrates with the tools and platforms that matter most to our users. We partner with industry leaders like Snowflake, Databricks, Google, and Ververica to ensure our streaming platform works seamlessly with data warehouses, lakehouse storage, and stream processing frameworks. Forrester’s evaluation observed that StreamNative’s ",[36,161,162],{},"\"investments in Pulsar’s open-source ecosystem and performance optimization make it the primary platform for enterprises wishing to implement Pulsar.\"",[40,164,166],{"id":165},"powering-real-time-use-cases-across-industries",[44,167,168],{},"Powering Real-Time Use Cases Across Industries",[48,170,171],{},"One of the greatest validations of StreamNative's approach is the success our customers are achieving with real-time data. StreamNative's platform is versatile and use-case agnostic—if an application demands high-volume, low-latency data movement, we can power it. This flexibility is why our customer base spans industries from finance and IoT to major automobile manufacturers and online gaming. The common thread is that these organizations need to process and react to data in milliseconds, and StreamNative is delivering the capabilities to make that possible.",[48,173,174],{},"Cisco uses StreamNative to underpin an IoT telemetry system of colossal scale, connecting hundreds of millions of devices and thousands of enterprise clients with real-time data streams. The platform's multi-tenant design and proven reliability allow Cisco to offer its customers a live feed of device data with unwavering confidence. In the financial sector, FICO has built streaming pipelines on StreamNative to detect fraud as transactions happen and to monitor systems in real time. With StreamNative's strong guarantees around message durability and ordering, FICO can catch anomalies or suspicious patterns within seconds. And in digital customer engagement, Iterable relies on StreamNative to process billions of events every day—clicks, views, purchases—so that marketers can trigger personalized campaigns instantly based on user behavior.",[48,176,177],{},"Our customers uniformly deal with mission-critical data streams, where downtime or delays are unacceptable. StreamNative's fault-tolerant, scalable infrastructure has proven equal to the task, handling scenarios like bursting to millions of events per second or seamlessly spanning multiple cloud regions. Forrester's report recognized StreamNative for supporting event-driven architectures with robust scalability—which for us is a reflection of our platform's ability to meet the most demanding enterprise requirements.",[40,179,181],{"id":180},"continuing-to-innovate-ursa-orca-and-the-road-ahead",[44,182,183],{},"Continuing to Innovate: Ursa, Orca, and the Road Ahead",[48,185,186,187,190],{},"While we are thrilled to be recognized in Forrester's Streaming Data Platforms Wave, we view this as just the beginning. StreamNative's vision has always been bold: to ",[44,188,189],{},"provide a unified platform that not only handles today's streaming needs but also anticipates the emerging requirements of tomorrow",".",[48,192,193],{},"One key area of focus is the convergence of streaming data with advanced analytics and AI. As Forrester points out in the report, technology leaders should look for platforms that natively integrate messaging, stream processing, and analytics to provide AI agents with real-time, contextualized information. We couldn't agree more. Our award-winning Ursa Engine and Orca Agent Engine are aimed at extending our platform up the stack—bridging the gap between data streams and data lakes, and between event streams and intelligent processing.",[48,195,196],{},"Our new Ursa Engine introduces a lakehouse-native approach to streaming: it can write events directly to table formats like Iceberg on cloud storage, eliminating entire classes of ETL jobs and making fresh data instantly available for analytics queries. By integrating streaming and lakehouse technologies, we help customers collapse data silos and accelerate their AI\u002FML pipelines.",[48,198,199,200,203],{},"Beyond analytics integration, we are also enhancing StreamNative with more out-of-the-box processing and governance capabilities. In the coming months, we plan to introduce new features for lightweight stream processing and transformation, making it easier to build reactive applications directly on the platform. We're also expanding our ecosystem of connectors and integrations, so that whether your data lands in Snowflake, Databricks, or an AI model, StreamNative will seamlessly feed it. ",[44,201,202],{},"We're investing significantly in enterprise features including security, schema registry, governance, and monitoring tooling","—capabilities that are essential for mission-critical deployments and where we're committed to continued improvement.",[48,205,206],{},"This recognition from Forrester energizes us to keep innovating at full speed. We're sharing this honor with our amazing customers, community, and partners who drive us forward every day. Your feedback and real-world challenges have helped shape StreamNative into what it is today, and together, we will shape the future of streaming data. Thank you for joining us on this journey—we're just getting started, and we can't wait to deliver even more value as we continue to evolve our platform. Onward to real-time everything!",[208,209],"hr",{},[32,211,213],{"id":212},"streamnative-in-the-forrester-wave-evaluation-findings",[44,214,215,216,223],{},"StreamNative in ",[44,217,218],{},[55,219,220],{"href":57},[44,221,222],{},"The Forrester Wave™",": Evaluation Findings",[225,226,228],"h5",{"id":227},"recognized-as-a-contender-among-15-streaming-data-platform-providers","• Recognized as a Contender among 15 streaming data platform providers",[225,230,232],{"id":231},"received-the-highest-scores-possible-50-in-both-the-messaging-and-resource-optimization-criteria","* Received the highest scores possible (5.0) in both the Messaging and Resource Optimization criteria",[225,234,236],{"id":235},"cited-as-the-primary-platform-for-enterprises-wishing-to-implement-pulsar","• Cited as the primary platform for enterprises wishing to implement Pulsar",[225,238,240],{"id":239},"noted-for-excelling-at-messaging-and-resource-optimization","• Noted for excelling at messaging and resource optimization",[225,242,244],{"id":243},"customers-cited-lower-infrastructure-costs-and-excellent-support-responsiveness","• Customers cited lower infrastructure costs and excellent support responsiveness",[225,246,248],{"id":247},"recognized-for-supporting-event-driven-architectures-with-robust-scalability","• Recognized for supporting event-driven architectures with robust scalability",[208,250],{},[252,253,255,256,259,260,190],"h6",{"id":254},"forrester-disclaimer-forrester-does-not-endorse-any-company-product-brand-or-service-included-in-its-research-publications-and-does-not-advise-any-person-to-select-the-products-or-services-of-any-company-or-brand-based-on-the-ratings-included-in-such-publications-information-is-based-on-the-best-available-resources-opinions-reflect-judgment-at-the-time-and-are-subject-to-change-for-more-information-read-about-forresters-objectivity-here","**Forrester Disclaimer: **",[36,257,258],{},"Forrester does not endorse any company, product, brand, or service included in its research publications and does not advise any person to select the products or services of any company or brand based on the ratings included in such publications. Information is based on the best available resources. Opinions reflect judgment at the time and are subject to change",". *For more information, read about Forrester’s objectivity *",[55,261,265],{"href":262,"rel":263},"https:\u002F\u002Fwww.forrester.com\u002Fabout-us\u002Fobjectivity\u002F",[264],"nofollow",[36,266,267],{},"here",[208,269],{},[252,271,273],{"id":272},"apache-apache-pulsar-apache-kafka-apache-flink-and-other-names-are-trademarks-of-the-apache-software-foundation-no-endorsement-by-apache-or-other-third-parties-is-implied",[36,274,275],{},"Apache®, Apache Pulsar®, Apache Kafka®, Apache Flink® and other names are trademarks of The Apache Software Foundation. No endorsement by Apache or other third parties is implied.",{"title":18,"searchDepth":19,"depth":19,"links":277},[278,280,281,282,283,284,285],{"id":34,"depth":279,"text":38},3,{"id":42,"depth":19,"text":46},{"id":83,"depth":19,"text":86},{"id":111,"depth":19,"text":114},{"id":136,"depth":19,"text":139},{"id":165,"depth":19,"text":168},{"id":180,"depth":19,"text":183,"children":286},[287],{"id":212,"depth":279,"text":288},"StreamNative in The Forrester Wave™: Evaluation Findings",null,"Company","2025-12-16","StreamNative is recognized in The Forrester Wave™: Streaming Data Platforms, Q4 2025. Discover why Forrester highlights StreamNative's high-performance messaging, efficient resource use, and cost-effective Kafka API compatibility for real-time innovation.","\u002Fimgs\u002Fblogs\u002F693bd36cf01b217dcb67278f_Streamnative_blog_thumbnail.png",false,{},0,"\u002Fblog\u002Fstreamnative-recognized-in-the-forrester-wave-streaming-data-platforms-2025","10 mins read",{"title":26,"description":292},"blog\u002Fstreamnative-recognized-in-the-forrester-wave-streaming-data-platforms-2025",[302,303,304],"Announcements","Real-Time","Forrester","5Nr1vAcqlQ7yFQfdL0a3MLsNFerVmEOQJXD9Twz5lx8",{"id":307,"title":308,"authors":309,"body":314,"canonicalUrl":289,"category":1073,"createdAt":289,"date":1074,"description":1075,"extension":8,"featured":7,"image":1076,"isDraft":294,"link":289,"meta":1077,"navigation":7,"order":296,"path":1078,"readingTime":1079,"relatedResources":289,"seo":1080,"stem":1081,"tags":1082,"__hash__":1085},"blogs\u002Fblog\u002Fhow-we-run-a-5-gb-s-kafka-workload-for-just-50-per-hour.md","How We Run a 5 GB\u002Fs Kafka Workload for Just $50 per Hour",[310,311,312,313],"Matteo Meril","Neng Lu","Hang Chen","Penghui Li",{"type":15,"value":315,"toc":1043},[316,319,322,325,328,331,335,338,348,354,357,365,370,374,381,384,387,395,399,402,407,411,414,417,420,423,432,436,439,450,453,457,460,463,474,477,481,485,493,496,500,508,537,541,544,549,553,556,560,563,566,571,580,585,588,591,602,606,609,620,624,627,630,635,638,667,671,673,679,682,687,692,695,699,713,717,728,732,747,756,767,770,773,777,780,783,794,797,800,803,808,813,817,821,838,842,856,861,865,876,879,895,899,910,915,920,928,932,935,939,946,950,953,962,967,976,982,991,1000,1009,1018,1027,1035],[48,317,318],{},"The rise of DeepSeek has shaken the AI infrastructure market, forcing companies to confront the escalating costs of training and deploying AI models. But the real pressure point isn’t just compute—it’s data acquisition and ingestion costs.",[48,320,321],{},"As businesses rethink their AI cost-containment strategies, real-time data streaming is emerging as a critical enabler. The growing adoption of Kafka as a standard protocol has expanded cost-efficient options, allowing companies to optimize streaming analytics while keeping expenses in check.",[48,323,324],{},"Ursa, the data streaming engine powering StreamNative’s managed Kafka service, is built for this new reality. With its leaderless architecture and native lakehouse storage integration, Ursa eliminates costly inter-zone network traffic for data replication and client-to-broker communication while ensuring high availability at minimal operational cost.",[48,326,327],{},"In this blog post, we benchmarked the infrastructure cost and total cost of ownership (TCO) for running a 5GB\u002Fs Kafka workload across different Kafka vendors, including Redpanda, Confluent WarpStream, and AWS MSK. Our benchmark results show that Ursa can sustain 5GB\u002Fs Kafka workloads at just 5% of the cost of traditional streaming engines like Redpanda—making it the ideal solution for high-performance, cost-efficient ingestion and data streaming for data lakehouses and AI workloads.",[48,329,330],{},"Note: We also evaluated vanilla Kafka in our benchmark; however, for simplicity, we have focused our cost comparison on vendor solutions rather than self-managed deployments. That said, it is important to highlight that both Redpanda and vanilla Kafka use a leader-based data replication approach. In a data-intensive, network-bound workload like 5GB\u002Fs streaming, with the same machine type and replication factor, Redpanda and vanilla Kafka produced nearly identical cost profiles.",[40,332,334],{"id":333},"key-benchmark-findings","Key Benchmark Findings",[48,336,337],{},"Ursa delivered 5 GB\u002Fs of sustained throughput at an infrastructure cost of just $54 per hour. For comparison:",[339,340,341,345],"ul",{},[342,343,344],"li",{},"MSK: $303 per hour → 5.6x more expensive compared to Ursa",[342,346,347],{},"Redpanda: $988 per hour → 18x more expensive compared to Ursa",[48,349,350],{},[351,352],"img",{"alt":18,"src":353},"\u002Fimgs\u002Fblogs\u002F679c71b67d9046f26edc7977_AD_4nXfvTqyBNUBu2lObdkKAx-5UNkpNP8UYULLZyOcixE6z99VMZUUEsUqWjzexI7vjyNGRNSAUoM9smYvdTP55ctAhIbrs5lmQgcSVMWdaoigbWouCl95DVSQsxooY-qqfGcYqS4g4zA.png",[48,355,356],{},"Beyond infrastructure costs, when factoring in both storage pricing, vendor pricing and operational expenses, Ursa’s total cost of ownership (TCO) for a 5GB\u002Fs workload with a 7-day retention period is:",[339,358,359,362],{},[342,360,361],{},"50% cheaper than Confluent WarpStream",[342,363,364],{},"85% cheaper than MSK and Redpanda",[48,366,367],{},[351,368],{"alt":18,"src":369},"\u002Fimgs\u002Fblogs\u002F679c602d77e9c706de5343b8_AD_4nXeDv8rrv_C1CTCCiqYo1zpvlGYbdBk1r0VEqovAPu22iFMQZgh54Hfw9PBMLzM7jDFxKwAFDxbdG0np4XVk_tGsWhEKMloLRcmmea7lvueCx-0cFsyaE3Mya4Mxc1Dox95A6JEc.png",[40,371,373],{"id":372},"ursa-highly-cost-efficient-data-streaming-at-scale","Ursa: Highly Cost-Efficient Data Streaming at Scale",[48,375,376,380],{},[55,377,379],{"href":378},"\u002Fblog\u002Fursa-reimagine-apache-kafka-for-the-cost-conscious-data-streaming","Ursa"," is a next-generation data streaming engine designed to deliver high performance at a fraction of the cost of traditional disk-based solutions. It is fully compatible with Apache Kafka and Apache Pulsar APIs, while leveraging a leaderless, lakehouse-native architecture to maximize scalability, efficiency, and cost savings.",[48,382,383],{},"Ursa’s key innovation is separating storage from compute and decoupling metadata\u002Findex operations from data operations by utilizing cloud object storage (e.g., AWS S3) instead of costly inter-zone disk-based replication. It also employs open lakehouse formats (Iceberg and Delta Lake), enabling columnar compression to significantly reduce storage costs while maintaining durability and availability.",[48,385,386],{},"In contrast, traditional streaming systems—like Kafka and Redpanda—depend on leader-based architectures, which drive up inter-zone traffic costs due to replication and client communication. Ursa mitigates these costs by:",[339,388,389,392],{},[342,390,391],{},"Eliminating inter-zone traffic costs via a leaderless architecture.",[342,393,394],{},"Replacing costly inter-zone replication with direct writes to cloud storage using open lakehouse formats.",[40,396,398],{"id":397},"how-ursa-eliminates-inter-zone-traffic","How Ursa Eliminates Inter-Zone Traffic",[48,400,401],{},"Ursa minimizes inter-zone traffic by leveraging a leaderless architecture, which eliminates inter-zone communication between clients and brokers, and lakehouse-native storage, which removes the need for inter-zone data replication. This approach ensures high availability and scalability while avoiding unnecessary cross-zone data movement.",[48,403,404],{},[351,405],{"alt":18,"src":406},"\u002Fimgs\u002Fblogs\u002F679c602e21b3571bb7117dca_AD_4nXd7Oahc77NjRLNvA9clLt0tsyU6MrIqVibFYv5pW5giTIcCHPr3EA_yTGzfVEUIVO3VXK56qWK8zmBCp5lY0E_4nmlWIPFrHjtHylA5NhwELjn-UB0fLG2h_kbrxrc7Cs_edvveNA.png",[32,408,410],{"id":409},"leaderless-architecture","Leaderless architecture",[48,412,413],{},"Traditional streaming engines such as Kafka, Pulsar, or RedPanda rely on a leader-based model, where each partition is assigned to a single leader broker that handles all writes and reads.",[48,415,416],{},"Pros of Leader-Based Architectures:\n✔ Maintains message ordering via local sequence IDs\n✔ Delivers low latency and high performance through message caching",[48,418,419],{},"Cons of Leader-Based Architectures:\n✖ Throughput bottlenecked by a single broker per partition\n✖ Inter-zone traffic required for high availability in multi-AZ deployments",[48,421,422],{},"While Kafka and Pulsar offer partial solutions (e.g., reading from followers, shadow topics) to reduce read-related inter-zone traffic, producers still send data to a single leader.",[48,424,425,426,431],{},"Ursa removes the concept of topic ownership, allowing any broker in the cluster to handle reads or writes for any partition. The primary challenge—ensuring message ordering—is solved with ",[55,427,430],{"href":428,"rel":429},"https:\u002F\u002Fgithub.com\u002Fstreamnative\u002Foxia",[264],"Oxia",", a scalable metadata and index service created by StreamNative in 2022.",[32,433,435],{"id":434},"oxia-the-metadata-layer-enabling-leaderless-architecture","Oxia: The Metadata Layer Enabling Leaderless Architecture",[48,437,438],{},"Ensuring message ordering in a leaderless architecture is complex, but Ursa solves this with Oxia:",[339,440,441,444,447],{},[342,442,443],{},"Handles millions of metadata\u002Findex operations per second",[342,445,446],{},"Generates sequential IDs to maintain strict message ordering",[342,448,449],{},"Optimized for Kubernetes with horizontal scalability",[48,451,452],{},"Producers and consumers can connect to any broker within their local AZ, eliminating inter-zone traffic costs while maintaining performance through localized caching.",[32,454,456],{"id":455},"zero-interzone-data-replication","Zero interzone data replication",[48,458,459],{},"In most distributed systems, data replication from a leader (primary) to followers (replicas) is crucial for fault tolerance and availability. However, replication across zones can inflate infrastructure expenses substantially.",[48,461,462],{},"Ursa avoids these costs by writing data directly to cloud storage (e.g., AWS S3, Google GCS):",[339,464,465,468,471],{},[342,466,467],{},"Built-In Resilience: Cloud storage inherently offers high availability and fault tolerance without inter-zone traffic fees.",[342,469,470],{},"Tradeoff: Slightly higher latency (sub-second, with p99 at 500 milliseconds) compared to local disk\u002FEBS (single-digit to sub-100 milliseconds), in exchange for significantly lower costs (up to 10x lower).",[342,472,473],{},"Flexible Modes: Ursa is an addition to the classic BookKeeper-based engine, providing users with the flexibility to optimize for either cost or low latency based on their workload requirements.",[48,475,476],{},"By foregoing conventional replication, Ursa slashes inter-zone traffic costs and associated complexities—making it a compelling option for organizations seeking to balance high-performance data streaming with strict budget constraints.",[40,478,480],{"id":479},"how-we-ran-a-5-gbs-test-with-ursa","How We Ran a 5 GB\u002Fs Test with Ursa",[32,482,484],{"id":483},"ursa-cluster-deployment","Ursa Cluster Deployment",[339,486,487,490],{},[342,488,489],{},"9 brokers across 3 availability zones, each on m6i.8xlarge (Fixed 12.5 Gbps bandwidth, 32 vCPU cores, 128 GB memory).",[342,491,492],{},"Oxia cluster (metadata store) with 3 nodes of m6i.8xlarge, distributed across three availability zones (AZs).",[48,494,495],{},"During peak throughput (5 GB\u002Fs), each broker’s network usage was about 10 Gbps.",[32,497,499],{"id":498},"openmessaging-benchmark-workers-configuration","OpenMessaging Benchmark Workers & Configuration",[48,501,502,503,507],{},"The OpenMessaging Benchmark(OMB) Framework is a suite of tools that make it easy to benchmark distributed messaging systems in the cloud. Please check ",[55,504,505],{"href":505,"rel":506},"https:\u002F\u002Fopenmessaging.cloud\u002Fdocs\u002Fbenchmarks\u002F",[264]," for details.",[339,509,510,525,534],{},[342,511,512,513,518,519,524],{},"12 OMB workers: 6 for ",[55,514,517],{"href":515,"rel":516},"https:\u002F\u002Fgist.github.com\u002Fcodelipenghui\u002Fd1094122270775e4f1580947f80c5055",[264],"producers",", 6 for ",[55,520,523],{"href":521,"rel":522},"https:\u002F\u002Fgist.github.com\u002Fcodelipenghui\u002F06bada89381fb77a7862e1b4c1d8963d",[264],"consumers"," across 3 availability zones, on m6i.8xlarge instances. Each worker is configured with 12 CPU cores and 48 GB memory.",[342,526,527,528,533],{},"Sample YAML ",[55,529,532],{"href":530,"rel":531},"https:\u002F\u002Fgist.github.com\u002Fcodelipenghui\u002F204c1f26c4d44a218ae235bf2de99904",[264],"scripts"," provided for Kafka-compatible configuration and rate limits.",[342,535,536],{},"Achieved consistent 5 GB\u002Fs publish\u002Fsubscribe throughput.",[40,538,540],{"id":539},"ursa-benchmark-tests-results","Ursa Benchmark Tests & Results",[48,542,543],{},"The following diagram demonstrates that Ursa can consistently handle 5 GB\u002Fs of traffic, fully saturating the network across all broker nodes.",[48,545,546],{},[351,547],{"alt":18,"src":548},"\u002Fimgs\u002Fblogs\u002F679c602d7b261bac1113f7d6_AD_4nXdDPsRc3koXICiFF0bqSmGWbJt_RlUy4FE3ruuWOfbCfpcqZ1dejjqGbkaCJv2hQFL1nirRouBVRW2l5uMWBvY9naMqGB_wHcLI14dBM0f85TXhmdm3UxEv1yGX9Y4hf5FttSkZew.png",[40,550,552],{"id":551},"comparing-infrastructure-cost","Comparing Infrastructure Cost",[48,554,555],{},"This benchmark first evaluates infrastructure costs of running a 5 GB\u002Fs streaming workload (1:1 producer-to-consumer ratio) across different data streaming engines, including Ursa, Redpanda, and AWS MSK, with a focus on multi-AZ deployments to ensure a fair comparison.",[32,557,559],{"id":558},"test-setup-key-assumptions","Test Setup & Key Assumptions",[48,561,562],{},"All tests use multi-AZ configurations, with clusters and clients distributed across three AWS availability zones (AZs). Cluster size scales proportionally to the number of AZs, and rack-awareness is enabled for all engines to evenly distribute topic partitions and leaders.",[48,564,565],{},"To ensure a fair comparison, we selected the same machine type capable of fully utilizing both network and storage bandwidth for Ursa and Redpanda in this 5GB\u002Fs test:",[339,567,568],{},[342,569,570],{},"9 × m6i.8xlarge instances",[48,572,573,574,579],{},"However, MSK's storage bandwidth limits vary depending on the selected instance type, with the highest allowed limit capped at 1000 MiB\u002Fs per broker, according to",[55,575,578],{"href":576,"rel":577},"https:\u002F\u002Fdocs.aws.amazon.com\u002Fmsk\u002Flatest\u002Fdeveloperguide\u002Fmsk-provision-throughput-management.html#throughput-bottlenecks",[264]," AWS documentation",". Given this constraint, achieving 5 GB\u002Fs throughput with a replication factor of 3 required the following setup:",[339,581,582],{},[342,583,584],{},"15 × kafka.m7g.8xlarge (32 vCPUs, 128 GB memory, 15 Gbps network, 4000 GiB EBS).",[48,586,587],{},"This configuration was necessary to work around MSK's storage bandwidth limitations, ensuring a comparable cost basis to other evaluated streaming engines.",[48,589,590],{},"Additional key assumptions include:",[339,592,593,596,599],{},[342,594,595],{},"Inter-AZ producer traffic: For leader-based engines, two-thirds of producer-to-broker traffic crosses AZs due to leader distribution.",[342,597,598],{},"Consumer optimizations: Follower fetch is enabled across all tests, eliminating inter-AZ consumer traffic.",[342,600,601],{},"Storage cost exclusions: This benchmark only evaluates streaming costs, assuming no long-term data retention.",[32,603,605],{"id":604},"inter-broker-replication-costs","Inter-Broker Replication Costs",[48,607,608],{},"Inter-broker (cross-AZ) replication is a major cost driver for data streaming engines:",[339,610,611,614,617],{},[342,612,613],{},"RedPanda: Inter-broker replication is not free, leading to substantial costs when data must be copied across multiple availability zones.",[342,615,616],{},"AWS MSK: Inter-broker replication is free, but MSK instance pricing is significantly higher (e.g., $3.264 per hour for kafka.m7g.8xlarge vs $1.306 per hour for an on-demand m7g.8xlarge). The storage price of MSK is $0.10 per GB-month which is significantly higher than st1, which costs $0.045 per GB-month. Even though replication is free, client-to-broker traffic still incurs inter-AZ charges.",[342,618,619],{},"Ursa: No inter-broker replication costs due to its leaderless architecture, eliminating inter-zone replication costs entirely.",[32,621,623],{"id":622},"zone-affinity-reducing-inter-az-costs","Zone Affinity: Reducing Inter-AZ Costs",[48,625,626],{},"We evaluated zone affinity mechanisms to further reduce inter-AZ data transfer costs.",[48,628,629],{},"Consumers:",[339,631,632],{},[342,633,634],{},"Follower fetch is enabled across all tests, ensuring consumers fetch data from replicas in their local AZ—eliminating inter-zone consumer traffic except for metadata lookups",[48,636,637],{},"Producers:",[339,639,640,649,658],{},[342,641,642,643,648],{},"Kafka protocol lacks an easy way to enforce producer AZ affinity (though ",[55,644,647],{"href":645,"rel":646},"https:\u002F\u002Fcwiki.apache.org\u002Fconfluence\u002Fdisplay\u002FKAFKA\u002FKIP-1123:+Rack-aware+partitioning+for+Kafka+Producer",[264],"KIP-1123"," aims to address this). And it only works with the default partitioner (i.e., when no record partition or record key is specified).",[342,650,651,652,657],{},"Redpanda recently introduced ",[55,653,656],{"href":654,"rel":655},"https:\u002F\u002Fdocs.redpanda.com\u002Fredpanda-cloud\u002Fdevelop\u002Fproduce-data\u002Fleader-pinning\u002F",[264],"leader pinning",", but this only benefits setups where producers are confined to a single AZ—not applicable to our multi-AZ benchmark.",[342,659,660,661,666],{},"Ursa is the only system in this test with ",[55,662,665],{"href":663,"rel":664},"https:\u002F\u002Fdocs.streamnative.io\u002Fdocs\u002Fconfig-kafka-client#eliminate-cross-az-networking-traffic",[264],"built-in zone affinity for both producers and consumers",". It achieves this by embedding producer AZ information in client.id, allowing metadata lookups to route clients to local-AZ brokers, eliminating inter-AZ producer traffic.",[32,668,670],{"id":669},"cost-comparison-results","Cost Comparison Results",[48,672,337],{},[339,674,675,677],{},[342,676,344],{},[342,678,347],{},[48,680,681],{},"Ursa’s leaderless architecture, zone affinity, and native cloud storage integration deliver unparalleled cost efficiency, making it the most cost-effective choice for high-throughput data streaming workloads.",[48,683,684],{},[351,685],{"alt":18,"src":686},"\u002Fimgs\u002Fblogs\u002F679c72208198ca36a352f228_AD_4nXeeZuM8T-xBlD4Vf3j67K618n08qh8wIDLLtiLJG0ssA1Wj1V26u7wIDTX9sqLrtw8mB2c299dwzarGen62CG0Vh7nWstn5qbPGFcBaKJYEepTsLr5fHWv1U8uqbg8Y0UOK6fJ7.png",[48,688,689],{},[351,690],{"alt":18,"src":691},"\u002Fimgs\u002Fblogs\u002F679c625978031f40229de484_AD_4nXdLkLLJ30KKr-_A_rN1j8akVwBYacAWIPzWHoOReJF421890kfByZoQQxkLczihVSmiw5Q9J51-V9I2SEKITbwsYnANDDTlAVL5nQ_jfaHNTe9VEWhSoa7DZooCnilDYL6l6msmJg.png",[48,693,694],{},"The detailed infrastructure cost calculations for each data streaming engine are listed below:",[32,696,698],{"id":697},"streamnative-ursa","StreamNative - Ursa",[339,700,701,704,707,710],{},[342,702,703],{},"Server EC2 costs: 9 * $1.536\u002Fhr = $14",[342,705,706],{},"Client EC2 costs: 9 * $1.536\u002Fhr =$14",[342,708,709],{},"S3 write requests costs: 1350 r\u002Fs * $0.005\u002F1000r * 3600s = $24",[342,711,712],{},"S3 read requests costs: 1350 r\u002Fs * $0.0004\u002F1000r * 3600s = $2",[32,714,716],{"id":715},"aws-msk","AWS MSK",[339,718,719,722,725],{},[342,720,721],{},"Server EC2 costs: 15 * $3.264\u002Fhr = $49",[342,723,724],{},"Client side EC2 costs: 9 * $1.536\u002Fhr =$14",[342,726,727],{},"Interzone traffic - producer to broker: 5GB\u002Fs * ⅔ * $0.02\u002FG(in+out) * 3600 = $240",[32,729,731],{"id":730},"redpanda","RedPanda",[339,733,734,736,738,741,744],{},[342,735,703],{},[342,737,706],{},[342,739,740],{},"Interzone traffic - producer to broker: 5GB\u002Fs * ⅔ * $0.02\u002FGB(in+out) * 3600 = $240",[342,742,743],{},"Interzone traffic - replication: 10GB\u002Fs * $0.02\u002FGB(in+out) * 3600 = $720",[342,745,746],{},"Interzone traffic - broker to consumer: $0 (fetch from local zone)",[48,748,749,750,755],{},"Please note that we were unable to test ",[55,751,754],{"href":752,"rel":753},"https:\u002F\u002Fwww.redpanda.com\u002Fblog\u002Fcloud-topics-streaming-data-object-storage",[264],"Redpanda with Cloud Topics",", as it remains an announced but unreleased feature and is not yet available for evaluation. Based on the limited information available, while Cloud Topics may help optimize inter-zone data replication costs, producers still need to traverse inter-availability zones to connect to the topic partition owners and incur inter-zone traffic costs of up to $240 per hour.",[339,757,758,764],{},[342,759,760,763],{},[55,761,647],{"href":645,"rel":762},[264]," (when implemented) will help mitigate producer-to-broker inter-zone traffic, but it is not yet available. And it only works with the default partitioner (no record partition or key is specified).",[342,765,766],{},"Redpanda’s leader pinning helps only when all producers for the pinned topic are confined to a single AZ. In multi-AZ environments (like our benchmark), inter-zone producer traffic remains unavoidable.",[48,768,769],{},"Additionally, Redpanda’s Cloud Topics architecture is not documented publicly. Their blog mentions \"leader placement rules to optimize produce latency and ingress cost,\" but it is unclear whether this represents a shift away from a leader-based architecture or if it uses techniques similar to Ursa’s zone-aware approach.",[48,771,772],{},"We may revisit this comparison as more details become available.",[40,774,776],{"id":775},"comparing-total-cost-of-ownership","Comparing Total Cost of Ownership",[48,778,779],{},"As highlighted earlier, with a BYOC Ursa setup, you can achieve 5 GB\u002Fs throughput at just 5% of the infrastructure cost of a traditional leader-based data streaming engine, such as Kafka or RedPanda, while managing the infrastructure yourself. This significant cost reduction is enabled by Ursa’s leaderless architecture and lakehouse-native storage design, which eliminate overhead costs such as inter-zone traffic and leader-based data replication. By leveraging a lakehouse-native, leaderless architecture, Ursa reduces resource requirements, enabling you to handle high data throughput efficiently and at a fraction of the cost of RedPanda.",[48,781,782],{},"Now, let’s examine the total cost comparison, evaluating Ursa alongside other vendors, including those that have adopted a leaderless architecture (e.g., Confluent WarpStream). This comparison is based on a 5GB\u002Fs workload with a 7-day retention period, factoring in both storage cost and vendor costs Here are the key findings:",[339,784,785,788,791],{},[342,786,787],{},"Ursa ($164,353\u002Fmonth) is: 50% cheaper than Confluent WarpStream ($337,068\u002Fmonth)",[342,789,790],{},"85% cheaper than AWS MSK ($1,115,251\u002Fmonth)",[342,792,793],{},"86% cheaper than Redpanda ($1,202,853\u002Fmonth)",[48,795,796],{},"In addition to Ursa’s architectural advantages—eliminating most inter-AZ traffic and leveraging lakehouse storage for cost-effective data retention—it also adopts a more fair and cost-efficient pricing model: Elastic Throughput-based pricing. This approach aligns costs with actual usage, avoiding unnecessary overhead.",[48,798,799],{},"Unlike WarpStream, which charges for both storage and throughput, Ursa ensures that customers only pay for the throughput they actively use. Ursa’s pricing is based on compressed data sent by clients, meaning the more data compressed on the client side, the lower the cost. In contrast, WarpStream prices are based on uncompressed data, unfairly inflating expenses and failing to incentivize customers to optimize their client applications.",[48,801,802],{},"This distinction is crucial, as compressed data reduces both storage and network costs, making Ursa’s pricing model not only more cost-effective but also more transparent and predictable.",[48,804,805],{},[351,806],{"alt":18,"src":807},"\u002Fimgs\u002Fblogs\u002F679c602d194800c9206d9d58_AD_4nXcFlf755xgyz7htxhMhBV5fGrsxy642mQNodt61DTok_z1dwkw5A6lkO5hatXVneCaB0anbZPAyvLI3MlIMuQEYLEACHHvQMOr5UfaB37dfzkdqewDEvcT-20VGd_zzvJsuA00zGA.png",[48,809,810],{},[351,811],{"alt":18,"src":812},"\u002Fimgs\u002Fblogs\u002F679c62594e9c2e629fae73aa_AD_4nXeU6cOgItnjLsEZCOf13TEvMY_SHWWIxYP2OYUj-B1GUPyWO78OG08K_v03hwYSVcg06f9dqDiGmdwy76vynjmiDGL5bluZ5_XF4nSU_r59oOZdfViXndXt6s11vVOY7qwfZN8v.png",[32,814,816],{"id":815},"cost-breakdown","Cost Breakdown",[818,819,820],"h4",{"id":697},"StreamNative – Ursa",[339,822,823,826,829,832,835],{},[342,824,825],{},"EC2 (Server): 9 × $1.536\u002Fhr × 24 hr × 30 days = $9,953.28",[342,827,828],{},"S3 Write Requests: 1,350 r\u002Fs × $0.005\u002F1,000 r × 3,600 s × 24 hr × 30 days = $17,496",[342,830,831],{},"S3 Read Requests: 1,350 r\u002Fs × $0.0004\u002F1,000 r × 3,600 s × 24 hr × 30 days = $1,400",[342,833,834],{},"S3 Storage Costs: 5 GB\u002Fs × $0.021\u002FGB × 3,600 s × 24 hr × 7 days = $63,504",[342,836,837],{},"Vendor Cost: 200 ETU × $0.50\u002Fhr × 24 hr × 30 days = $72,000",[818,839,841],{"id":840},"warpstream","WarpStream",[339,843,844,847],{},[342,845,846],{},"Based on WarpStream’s pricing calculator (as of January 29, 2025), we assume a 4:1 client data compression ratio, meaning 20 GB\u002Fs of uncompressed data translates to 5 GB\u002Fs of compressed data.",[342,848,849,850,855],{},"It's important to note that WarpStream’s pricing structure has fluctuated frequently throughout January. We observed the cost reported by their calculator changing from $409,644 per month to $337,068 per month. This variability has been previously highlighted in the blog post “",[55,851,854],{"href":852,"rel":853},"https:\u002F\u002Fbigdata.2minutestreaming.com\u002Fp\u002Fthe-brutal-truth-about-apache-kafka-cost-calculators",[264],"The Brutal Truth About Kafka Cost Calculators","”. To ensure transparency, we have documented the pricing as of January 29, 2025.",[48,857,858],{},[351,859],{"alt":18,"src":860},"\u002Fimgs\u002Fblogs\u002F679c602e42713e0028e9af5e_AD_4nXcu5_VWTLu9jRYs6zX1MBAOtLQEo5gyfNSWPcbpnQHXTa8qNCFAXezRR2E8daygzYTTwd4dhJjaLaLM8C6y_3OGbu2NS7pdvEv3a8-ptNKOg7AeKnYqPQCAYvQ5EuxzuI3JYIvY.png",[818,862,864],{"id":863},"msk","MSK",[339,866,867,870,873],{},[342,868,869],{},"EC2 (Server): 15 * $3.264\u002Fhr × 24 hr × 30 days = $35,251",[342,871,872],{},"Interzone Traffic (Client-Server): 5 GB\u002Fs × ⅔ × $0.02\u002FGB (in+out) × 3,600 s × 24 hr × 30 days = $172,800",[342,874,875],{},"Storage: 5 GB\u002Fs × $0.1\u002FGB-month × 3,600 s × 24 hr × 7 days * 3 replicas = $907,200",[818,877,731],{"id":878},"redpanda-1",[339,880,881,884,886,889,892],{},[342,882,883],{},"EC2 (Server): 9 × $1.536\u002Fhr × 24 hr × 30 days = $9953",[342,885,872],{},[342,887,888],{},"Interzone Traffic (Replication): 5 GB\u002Fs × 2 × $0.02\u002FGB (in+out) × 3,600 s × 24 hr × 30 days = $518,400",[342,890,891],{},"Storage: 5 GB\u002Fs × $0.045\u002FGB-month(st1) × 3,600 s × 24 hr × 7 days * 3 replicas = $408,240",[342,893,894],{},"Vendor Cost: $93,333 per month (based on limited information. See additional notes below).",[818,896,898],{"id":897},"additional-notes","Additional Notes",[339,900,901],{},[342,902,903,904,909],{},"Redpanda does not publicly disclose its BYOC pricing, making it difficult to accurately assess its total costs. We refer to information from the whitepaper “",[55,905,908],{"href":906,"rel":907},"https:\u002F\u002Fwww.redpanda.com\u002Fresources\u002Fredpanda-vs-confluent-performance-tco-benchmark-report#form",[264],"Redpanda vs. Confluent: A Performance and TCO Benchmark Report by McKnight Consulting Group.","” for estimation purposes. Based on the Tier-8 pricing model in the whitepaper,  the estimated cost to support a 5GB\u002Fs workload would be $1.12 million per year ($93,333 per month). However, since this calculation is based on an estimation, we will revisit and refine the cost assessment once Redpanda publishes its BYOC pricing.",[48,911,912],{},[351,913],{"alt":18,"src":914},"\u002Fimgs\u002Fblogs\u002F679c602dc8a9859eed89a0ef_AD_4nXdbcO8vsNNPy4GtkNLlmNKf22fjxRvzLzH7CtOna1L08sTbvnZx3HhufeFqc1w4K2gEF7lxO2IR5supotxebAiGnA07Qa8Yr3Rd1pVK2LYKK4WurlJGwgdwwucZIFoF-N_2oBjY.png",[48,916,917],{},[351,918],{"alt":18,"src":919},"\u002Fimgs\u002Fblogs\u002F679c602d6bc1c2287e012540_AD_4nXfcHZnLfjbjIr3ZAgoQXT9dwP3aQCOQPmGZZJUtpNZSwE6qY6M3yehIaBxCwxEIeu5PVdUPY0zhyjnow26YfgjdYgSG4GnV9ibxu0YWTIpwng6z_F6FUGJMpERMKtpsFESzXSN_Sw.png",[339,921,922,925],{},[342,923,924],{},"When estimating the storage costs for Kafka and Redpanda, we assume the use of HDD storage at $0.045\u002FGB, based on the premise that both systems can fully utilize disk bandwidth without incurring the higher costs associated with GP2 or GP3 volumes. However, in practice, many users opt for GP2 or GP3, significantly increasing the total storage cost for Kafka and Redpanda.",[342,926,927],{},"Unlike disk-based solutions, S3 storage does not require capacity preallocation—Ursa only incurs costs for the actual data stored. This contrasts with Kafka and Redpanda, where preallocating storage can drive up expenses. As a result, the real-world storage costs for Kafka and Redpanda are often 50% higher than the estimates above.",[40,929,931],{"id":930},"conclusion","Conclusion",[48,933,934],{},"Ursa represents a transformative shift in streaming data infrastructure, offering cost efficiency, scalability, and flexibility without compromising durability or reliability. By leveraging a leaderless architecture and eliminating inter-zone data replication, Ursa reduces total cost of ownership by over 90% compared to traditional leader-based streaming engines like Kafka and Redpanda. Its direct integration with cloud storage and scalable metadata & index management via Oxia ensure high availability and simplified infrastructure management.",[32,936,938],{"id":937},"balancing-latency-and-cost","Balancing Latency and Cost",[48,940,941,945],{},[55,942,944],{"href":943},"\u002Fblog\u002Fcap-theorem-for-data-streaming","Ursa trades off slightly higher latency for ultra low cost",", making it an ideal choice for the majority of streaming workloads, especially those that prioritize throughput and cost savings over ultra-low latency. Meanwhile, StreamNative’s BookKeeper-based engine remains the preferred solution for real-time, latency-sensitive applications. By combining these two approaches, StreamNative empowers customers with the flexibility to choose the right engine for their specific needs—whether it's maximizing cost savings or achieving ultra low-latency real-time performance.",[32,947,949],{"id":948},"the-future-of-streaming-infrastructure","The Future of Streaming Infrastructure",[48,951,952],{},"In an era where data fuels AI, analytics, and real-time decision-making, managing infrastructure costs is critical to sustaining innovation. Ursa is not just a cost-cutting alternative—it is a forward-thinking, lakehouse-native platform that redefines how modern data streaming infrastructure should be built and operated.",[48,954,955,956,961],{},"Whether your priority is reducing costs, improving flexibility, or ingesting massive data into lakehouses, Ursa delivers a future-proof solution for the evolving demands of real-time data streaming. ",[55,957,960],{"href":958,"rel":959},"https:\u002F\u002Fconsole.streamnative.cloud\u002F",[264],"Get started"," with StreamNative Ursa today!",[963,964,966],"h1",{"id":965},"references","References",[48,968,969,972,973],{},[970,971,430],"span",{}," ",[55,974,975],{"href":975},"\u002Fblog\u002Fintroducing-oxia-scalable-metadata-and-coordination",[48,977,978,972,980],{},[970,979,379],{},[55,981,378],{"href":378},[48,983,984,972,987],{},[970,985,986],{},"StreamNative pricing",[55,988,989],{"href":989,"rel":990},"https:\u002F\u002Fdocs.streamnative.io\u002Fdocs\u002Fbilling-overview",[264],[48,992,993,972,996],{},[970,994,995],{},"WarpStream pricing",[55,997,998],{"href":998,"rel":999},"https:\u002F\u002Fwww.warpstream.com\u002Fpricing#pricingfaqs",[264],[48,1001,1002,972,1005],{},[970,1003,1004],{},"AWS S3 pricing",[55,1006,1007],{"href":1007,"rel":1008},"https:\u002F\u002Faws.amazon.com\u002Fs3\u002Fpricing\u002F",[264],[48,1010,1011,972,1014],{},[970,1012,1013],{},"AWS EBS pricing",[55,1015,1016],{"href":1016,"rel":1017},"https:\u002F\u002Faws.amazon.com\u002Febs\u002Fpricing\u002F",[264],[48,1019,1020,972,1023],{},[970,1021,1022],{},"AWS MSK pricing",[55,1024,1025],{"href":1025,"rel":1026},"https:\u002F\u002Faws.amazon.com\u002Fmsk\u002Fpricing\u002F",[264],[48,1028,1029,972,1032],{},[970,1030,1031],{},"The Brutal Truth about Kafka Cost Calculators",[55,1033,852],{"href":852,"rel":1034},[264],[48,1036,1037,972,1040],{},[970,1038,1039],{},"Redpanda vs. Confluent: A Performance and TCO Benchmark Report by McKnight Consulting Group",[55,1041,906],{"href":906,"rel":1042},[264],{"title":18,"searchDepth":19,"depth":19,"links":1044},[1045,1046,1047,1052,1056,1057,1066,1069],{"id":333,"depth":19,"text":334},{"id":372,"depth":19,"text":373},{"id":397,"depth":19,"text":398,"children":1048},[1049,1050,1051],{"id":409,"depth":279,"text":410},{"id":434,"depth":279,"text":435},{"id":455,"depth":279,"text":456},{"id":479,"depth":19,"text":480,"children":1053},[1054,1055],{"id":483,"depth":279,"text":484},{"id":498,"depth":279,"text":499},{"id":539,"depth":19,"text":540},{"id":551,"depth":19,"text":552,"children":1058},[1059,1060,1061,1062,1063,1064,1065],{"id":558,"depth":279,"text":559},{"id":604,"depth":279,"text":605},{"id":622,"depth":279,"text":623},{"id":669,"depth":279,"text":670},{"id":697,"depth":279,"text":698},{"id":715,"depth":279,"text":716},{"id":730,"depth":279,"text":731},{"id":775,"depth":19,"text":776,"children":1067},[1068],{"id":815,"depth":279,"text":816},{"id":930,"depth":19,"text":931,"children":1070},[1071,1072],{"id":937,"depth":279,"text":938},{"id":948,"depth":279,"text":949},"StreamNative Cloud","2025-01-31","Discover how Ursa achieves 5GB\u002Fs Kafka workloads at just 5% of the cost of traditional streaming engines like Redpanda and AWS MSK. See our benchmark results comparing infrastructure costs, total cost of ownership (TCO), and performance across leading Kafka vendors.","\u002Fimgs\u002Fblogs\u002F679c6593d25099b1cdcec4ca_image-31.png",{},"\u002Fblog\u002Fhow-we-run-a-5-gb-s-kafka-workload-for-just-50-per-hour","30 min",{"title":308,"description":1075},"blog\u002Fhow-we-run-a-5-gb-s-kafka-workload-for-just-50-per-hour",[1083,1084,303],"TCO","Apache Kafka","CDUawvFKTs_AD8usvmIcTleU3mbfA0QAoPZM6xfVuo8",{"id":1087,"title":1088,"authors":1089,"body":1091,"canonicalUrl":289,"category":1589,"createdAt":289,"date":1590,"description":1591,"extension":8,"featured":294,"image":1592,"isDraft":294,"link":289,"meta":1593,"navigation":7,"order":296,"path":1594,"readingTime":1595,"relatedResources":289,"seo":1596,"stem":1597,"tags":1598,"__hash__":1599},"blogs\u002Fblog\u002Fhandling-100k-consumers-with-one-pulsar-topic.md","Handling 100K Consumers with One Pulsar Topic",[1090],"Hongjie Zhai",{"type":15,"value":1092,"toc":1570},[1093,1097,1106,1109,1112,1118,1121,1124,1128,1131,1134,1137,1154,1157,1168,1171,1175,1178,1181,1187,1207,1210,1216,1219,1223,1226,1232,1237,1240,1244,1247,1253,1256,1260,1263,1269,1272,1276,1279,1285,1306,1309,1312,1320,1324,1327,1341,1344,1348,1351,1357,1371,1374,1380,1384,1387,1398,1404,1407,1410,1414,1417,1420,1426,1429,1433,1436,1442,1445,1449,1452,1461,1471,1474,1477,1480,1483,1486,1492,1495,1501,1504,1510,1513,1515,1518,1521,1525,1567],[40,1094,1096],{"id":1095},"background","Background",[48,1098,1099,1100,1105],{},"Nippon Telegraph and Telephone Corporation (NTT) is one of the world's leading telecommunications carriers. ",[55,1101,1104],{"href":1102,"rel":1103},"https:\u002F\u002Fwww.rd.ntt\u002Fe\u002Fsic\u002F",[264],"NTT Software Innovation Center"," creates innovative platform technologies to support the ICT service for prosperous future as a professional group on IT. It works to create innovative software platforms and computing platform technologies to support the evolution of the IoT\u002FAI service as a professional group on IT. It will not only proactively contribute to the open source community but also promote research and development through open innovation. It will also contribute to the reduction of CAPEX\u002FOPEX for IT or strategic utilization of IT, using the accumulated technologies and know-how regarding software development and operation.",[48,1107,1108],{},"Before I introduce how we use Apache Pulsar to handle 100K consumers, let me first explain our use case and the challenges facing us.",[48,1110,1111],{},"In our smart city scenario, we need to collect data from a large number of devices, such as cars, sensors, and cameras, and further analyze the data for different purposes. For example, if a camera detects any road damage, we need to immediately broadcast the information to the cars nearby, thus avoiding traffic congestion. More specifically, we provide a topic for each area and all the vehicles in that area are connected to the topic. For a huge city, we expect that there are about 100K vehicles publishing data to a single topic. In addition to the large data volume, we also need to work with different protocols used by these devices, like MQTT, REST, and RTSP.",[48,1113,1114],{},[351,1115],{"alt":1116,"src":1117},"Visualization of how NTT collects data ","\u002Fimgs\u002Fblogs\u002F63be1482ae551659c9393d2c_NTT-blog-image1.png",[48,1119,1120],{},"Data persistence is another challenge in this scenario. For essential data, like key scenes from cameras or key events from IoT devices, we need to securely store them for further analysis, perhaps for a long period of time. We also have to prepare proper storage solutions in the system.",[48,1122,1123],{},"With massive devices, various protocols, and different storage systems, our data pipeline becomes extremely complicated. It is almost impossible to maintain such a huge system.",[40,1125,1127],{"id":1126},"why-did-we-choose-apache-pulsar","Why did we choose Apache Pulsar",[48,1129,1130],{},"As we worked on solutions, we were thinking about introducing a unified data hub, like a large, centralized message broker that is able to support various protocols. This way, all the devices only need to communicate with a single endpoint.",[48,1132,1133],{},"Nowadays, many brokers provide their own storage solutions or even support tiered storage, which guarantees persistence for any data processed by the brokers. This also means that we only need to work with brokers and their topics, which allows us to have an easier and cleaner system.",[48,1135,1136],{},"Ultimately, we chose to build our system with Apache Pulsar as the basic framework. Pulsar is a cloud-native streaming and messaging system with the following key features.",[339,1138,1139,1142,1151],{},[342,1140,1141],{},"A loosely-coupled architecture. Pulsar uses Apache BookKeeper as its storage engine. This allows us to independently scale out the storage cluster without changing the number of brokers if we need to store more data.",[342,1143,1144,1145,1150],{},"A pluggable protocol handler. Pulsar’s ",[55,1146,1149],{"href":1147,"rel":1148},"https:\u002F\u002Fgithub.com\u002Fapache\u002Fpulsar\u002Fblob\u002Fmaster\u002Fpulsar-broker\u002Fsrc\u002Fmain\u002Fjava\u002Forg\u002Fapache\u002Fpulsar\u002Fbroker\u002Fprotocol\u002FProtocolHandler.java",[264],"protocol handler"," enables us to work with multiple protocols with just one broadcaster. It supports MQTT, Kafka, and many other brokers. This makes it very convenient to ingest data from various sources into a centralized Pulsar cluster.",[342,1152,1153],{},"High performance and low latency. Pulsar shows excellent performance as we tested it using different benchmarks. We will talk about this in more detail later.",[48,1155,1156],{},"So, does Pulsar meet the performance requirements of our use case? Let’s take a look at the breakdown of our requirements.",[339,1158,1159,1162,1165],{},[342,1160,1161],{},"A large number of consumers. Brokers should be able to manage messages and broadcast them to up to 100K vehicles.",[342,1163,1164],{},"Low latency. We have tons of notifications generated against the data in real time, which need to be broadcast at an end-to-end (E2E) latency of less than 1 second. In our case, the end-to-end latency refers to the duration between the time a message is produced by cloud services and the time it is received by the vehicle. Technically, it contains two phases - producing and consuming.",[342,1166,1167],{},"Large messages. Brokers should be able to handle large messages from cameras (for example, video streams) without performance issues. Most brokers focus on handling small messages, such as event data from microservices on the cloud, which are usually about several hundred kilobytes at most. When messages become larger, these brokers may have performance problems.",[48,1169,1170],{},"In this blog, we will focus on the first 2 requirements, namely how to broadcast messages for 100K consumers with an end-to-end latency of less than 1 second.",[40,1172,1174],{"id":1173},"benchmark-testing","Benchmark testing",[48,1176,1177],{},"To understand how Pulsar fits into our use case, we performed some benchmark tests on Pulsar and I will introduce some of them in this section.",[48,1179,1180],{},"Figure 2 shows the general structure of our benchmark tests.",[48,1182,1183],{},[351,1184],{"alt":1185,"src":1186},"The structure of how NTT did their benchmark tests","\u002Fimgs\u002Fblogs\u002F63be152b5bc73f70a44008fa_NTT-blog-image2.png",[339,1188,1189,1192,1195,1198,1201,1204],{},[342,1190,1191],{},"Broadcast task: Only 1 publisher sending messages to 1 persistent topic with a single Pulsar broker",[342,1193,1194],{},"Consumers: 20K-100K consumers (shared subscription)",[342,1196,1197],{},"Message size: 10 KB",[342,1199,1200],{},"Message dispatch rate: 1 msg\u002Fs",[342,1202,1203],{},"Pulsar version: 2.10",[342,1205,1206],{},"Benchmark: OpenMessaging Benchmark Framework (OMB)",[48,1208,1209],{},"Figure 3 shows our client and cluster configurations.",[48,1211,1212],{},[351,1213],{"alt":1214,"src":1215},"Diagram of NTT's client and cluster configurations","\u002Fimgs\u002Fblogs\u002F63be1599f7dfe733e19dfc19_NTT-blog-image3.png",[48,1217,1218],{},"We performed the benchmark tests on Amazon Web Services (AWS), with both the broker and bookies using the same machine type (i3.4xlarge). We provided sufficient network (10 Gbit) and storage (2 SSDs) resources for each node to avoid hardware bottlenecks. This allowed us to focus on the performance of Pulsar itself. As we had too many consumers, we put them onto several servers, or clients in Figure 3.",[32,1220,1222],{"id":1221},"overall-benchmark-results","Overall benchmark results",[48,1224,1225],{},"Table 1 displays our benchmark results. We can see that Pulsar worked well with 20K consumers, recording a P99 latency of 0.68 seconds and a connection time of about 4 minutes. Both of them are acceptable in real-world usage.",[48,1227,1228],{},[351,1229],{"alt":1230,"src":1231},"Table of benchmark test results","\u002Fimgs\u002Fblogs\u002F63be1bf861fd137a60dda464_Screen-Shot-2023-01-10-at-6.16.06-PM.png",[339,1233,1234],{},[342,1235,1236],{},"Connection time: the time between the start of the connections to all consumers and the end of all the connections.",[48,1238,1239],{},"As the number of consumers increased, we noticed a decline in performance. When we had 30K consumers, the P99 latency exceeded 1 second. When 40K consumers were involved, the P99 latency even topped 4 seconds, with a connection time of nearly 20 minutes, which is too long for our use case. For 100K consumers, they even failed to establish the connections since they took too much time.",[32,1241,1243],{"id":1242},"a-polynomial-curve-the-connection-time-and-the-number-of-consumers","A polynomial curve: The connection time and the number of consumers",[48,1245,1246],{},"To understand how the connection time is related to consumers, we conducted further research and made a polynomial curve for the approximations of the collection time as the number of consumers increases.",[48,1248,1249],{},[351,1250],{"alt":1251,"src":1252},"Chart showing the connection time and the number of consumers ","\u002Fimgs\u002Fblogs\u002F63be1828096bb97d9e19b17d_NTT-blog-image4.png",[48,1254,1255],{},"Based on the curve, we expected the connection time to reach 8,000 seconds (about 2.2 hours) at 100K consumers, which is unacceptable for our case.",[32,1257,1259],{"id":1258},"connection-time-distribution-the-long-tail-problem","Connection time distribution: The long tail problem",[48,1261,1262],{},"In addition, for the case with 20K consumers, we measured the connection time of each consumer and created a histogram to see the time distribution across them, as depicted in Figure 5.",[48,1264,1265],{},[351,1266],{"alt":1267,"src":1268},"Histogram of connection time","\u002Fimgs\u002Fblogs\u002F63be185c61fd130442da3dac_NTT-blog-image5.png",[48,1270,1271],{},"The Y-axis represents the number of consumers that finished their connections within the time range on the X-axis. As shown in Figure 5, about 20% of connections finished in about 3 seconds, and more than half of the connections finished within one minute. The problem lay with the long tail. Some consumers even spent more than 200 seconds, which greatly affected the overall connection time.",[32,1273,1275],{"id":1274},"a-breakdown-of-p99-latency","A breakdown of P99 latency",[48,1277,1278],{},"For the P99 latency, we split it into six stages and measured their respective processing time in the 40K-consumer case.",[48,1280,1281],{},[351,1282],{"alt":1283,"src":1284},"Six stages of P99 latency for 40K consumers","\u002Fimgs\u002Fblogs\u002F63be18b2f7dfe75124a1131c_NTT-blog-image6.png",[1286,1287,1288,1291,1294,1297,1300,1303],"ol",{},[342,1289,1290],{},"Producing: Includes message production by the publisher, network communications, and protocol processing.",[342,1292,1293],{},"Broker internal process: Includes message deduplication, transformation, and other processes.",[342,1295,1296],{},"Message persistence: The communication between the broker and BookKeeper.",[342,1298,1299],{},"Notification: The broker receives an update notification from BookKeeper.",[342,1301,1302],{},"Broker internal process: The broker prepares the message for consumption.",[342,1304,1305],{},"Broadcasting: All the messages are broadcast to all the consumers.",[48,1307,1308],{},"Our results show that message persistence took up about 27% of the total latency while broadcasting accounted for about 33%. These two stages combined were responsible for most of the delay time, so we needed to focus on reducing the latency for them specifically.",[48,1310,1311],{},"Before I continue to explain how we worked out a solution, let’s review the conclusion of our benchmark results.",[1286,1313,1314,1317],{},[342,1315,1316],{},"Pulsar is already good enough for scenarios where there are no more than 20K consumers with a P99 latency requirement of less than 0.7s. The consumer connection time is also acceptable.",[342,1318,1319],{},"As the number of consumers increases, it takes more time for connections to finish. For 100K consumers, Pulsar still needs to be improved in terms of latency and connection time. For latency, the persistence (connections with BookKeeper) and broadcasting (connections with consumers & acks) stages take too much time.",[40,1321,1323],{"id":1322},"approaches-to-100k-consumers","Approaches to 100K consumers",[48,1325,1326],{},"There are typically two ways to improve performance: scale-up and scale-out. In our case, we can understand them in the following ways.",[339,1328,1329,1332],{},[342,1330,1331],{},"Scale-up: Improve the performance of a single broker.",[342,1333,1334,1335,1340],{},"Scale-out: Let multiple brokers handle one topic at the same time. One of the possible scale-out solutions is called “Shadow Topic”, proposed by a Pulsar PMC member. It allows us to distribute subscriptions across multiple brokers by creating \"copies\" of the original topic. See ",[55,1336,1339],{"href":1337,"rel":1338},"https:\u002F\u002Fgithub.com\u002Fapache\u002Fpulsar\u002Fissues\u002F16153",[264],"PIP-180"," for more details.",[48,1342,1343],{},"This blog will focus on the first approach. More specifically, we created a broadcast-specific model for better performance and resolved the task congestion issue when there are too many connections.",[32,1345,1347],{"id":1346},"four-subscription-types-in-pulsar","Four subscription types in Pulsar",[48,1349,1350],{},"First, let’s explore Pulsar’s subscription model. In fact, many brokers share similar models. In Pulsar, a topic must have at least one subscription to dispatch messages and each consumer must be linked to one subscription to receive messages. A subscription is responsible for transferring messages from topics. There are four types of subscriptions in Pulsar.",[48,1352,1353],{},[351,1354],{"alt":1355,"src":1356},"Diagram showing four subscription types in Pulsar","\u002Fimgs\u002Fblogs\u002F63be1900b949328cee8e2dcb_NTT-blog-image7.png",[339,1358,1359,1362,1365,1368],{},[342,1360,1361],{},"Exclusive. Only one consumer is allowed to be associated with the subscription. This means if the consumer crashes or disconnects, the messages in this subscription will not be processed anymore.",[342,1363,1364],{},"Failover. Supports multiple consumers, but only one of the consumers can receive messages. When the working consumer crashes or disconnects, Pulsar can switch to another consumer to make sure messages keep being processed.",[342,1366,1367],{},"Shared. Distributes messages across multiple consumers. Each consumer will only receive parts of the messages, and the number of messages will be well-balanced across every consumer.",[342,1369,1370],{},"Key_Shared. Similar to Shared subscriptions, Key_Shared subscriptions allow multiple consumers to be attached to the same subscription. Messages are delivered across consumers and the messages with the same key or same ordering key are sent to only one consumer.",[48,1372,1373],{},"A problem with the subscription types is that there is no model designed for sending the same messages to multiple consumers. This means in our broadcasting case, we must create a subscription for each consumer. As shown in Figure 8, for example, we used 4 exclusive subscriptions, and each of them had a connected consumer, allowing us to broadcast messages to all of them.",[48,1375,1376],{},[351,1377],{"alt":1378,"src":1379},"Diagram showing each exclusive subscription has a consumer attached","\u002Fimgs\u002Fblogs\u002F63be196408486ea2e5283b2b_NTT-blog-image8.png",[32,1381,1383],{"id":1382},"using-multiple-subscriptions-for-broadcasting-messages","Using multiple subscriptions for broadcasting messages",[48,1385,1386],{},"However, creating multiple subscriptions can increase latency, especially when you have too many consumers. To understand the reason, let’s take a look at how a subscription works. Figure 9 displays the general design of a subscription, which is comprised of three components:",[1286,1388,1389,1392,1395],{},[342,1390,1391],{},"The subscription itself.",[342,1393,1394],{},"Cursor. You use a cursor to track the position of consumers. You can consider it as a message ID, or the position on the message stream. This information will also be synchronized with the metadata store, which means you can resume consumption from this position even after the broker restarts.",[342,1396,1397],{},"Dispatcher. It is the only functional part of the subscription, which communicates with BookKeeper and checks if there are any new messages written to BookKeeper. If there are new messages, it will pull them out and send them to consumers.",[48,1399,1400],{},[351,1401],{"alt":1402,"src":1403},"Diagram of subscription components","\u002Fimgs\u002Fblogs\u002F63be19d6a5aeb58699adb203_NTT-blog-image9.png",[48,1405,1406],{},"As the dispatcher communicates with BookKeeper, each dispatch has its own connection to BookKeeper. This comes with a problem when you have too many consumers. In our case, 100K consumers were attached to 100K subscriptions, requiring 100K connections to BookKeeper. This huge number of connections was clearly a performance bottleneck.",[48,1408,1409],{},"In fact, these connections were redundant and unnecessary. This is because for this broadcasting task, all the consumers used their respective subscriptions just to retrieve the same messages from the same topic. Even for the cursor, as we sent the same data at the same time, we did not expect too many differences between these cursors. Theoretically, one cursor should be enough.",[32,1411,1413],{"id":1412},"broadcast-subscription-with-virtual-cursors","Broadcast Subscription with virtual cursors",[48,1415,1416],{},"To improve performance, we redesigned the subscription model specifically for handling large volumes of consumers (see Figure 10). The new structure guarantees the message order for each consumer. It shares many functions with the existing subscription model, such as cumulative acknowledgment.",[48,1418,1419],{},"In the new model, only one subscription exists to serve multiple consumers, which means there is only one dispatcher. As only a single connection to BookKeeper is allowed, this method can greatly reduce the load on BookKeeper and lower the latency. Additionally, since the subscription only has one cursor, there is no metadata duplication.",[48,1421,1422],{},[351,1423],{"alt":1424,"src":1425},"Figure of the new subscription model for the Broadcast Subscription","\u002Fimgs\u002Fblogs\u002F63be1a22b94932465c8ec2e4_NTT-blog-image10.png",[48,1427,1428],{},"In Pulsar, when consumers fail to receive or acknowledge messages, we need to resend the messages. To achieve this for one subscription and multiple consumers, we introduced a lightweight “virtual cursor” for each consumer to record the incremental position of the main cursor. The virtual cursor has a lightweight design; it does not contain any other information other than the incremental position. It allowed us to identify unread messages by comparing the virtual cursors and the data stored on BookKeeper. This way, we could keep unprocessed messages and delete any acknowledged ones.",[32,1430,1432],{"id":1431},"evaluating-the-performance-of-the-new-subscription-model","Evaluating the performance of the new subscription model",[48,1434,1435],{},"With this new subscription model, we evaluated its performance using 30K, 40K, and 100K consumers. The baseline is the shared subscription, which had the best result among all four original subscription models.",[48,1437,1438],{},[351,1439],{"alt":1440,"src":1441},"Table with benchmark test results of the Broadcast Subscription","\u002Fimgs\u002Fblogs\u002F63be1a725e78e6587ac87aa5_Screen-Shot-2023-01-10-at-6.09.40-PM.png",[48,1443,1444],{},"As shown in Table 2, when we had 40K consumers, the P99 latency of the Broadcast Subscription was almost 6 times faster than the original Shared Subscription. The connection time also saw a significant decrease as we only had one subscription. Even with 100K consumers, all the connections finished in just about 77.3 seconds. Although the results were extremely impressive, we still wanted a better P99 latency of less than 1 second.",[32,1446,1448],{"id":1447},"optimizing-orderedexecutor","Optimizing OrderedExecutor",[48,1450,1451],{},"In our benchmark evaluation, we found another factor that could lead to high latency: OrderedExecutor.",[48,1453,1454,1455,1460],{},"Let’s first explore how ",[55,1456,1459],{"href":1457,"rel":1458},"https:\u002F\u002Fbookkeeper.apache.org\u002Fdocs\u002Flatest\u002Fapi\u002Fjavadoc\u002Forg\u002Fapache\u002Fbookkeeper\u002Fcommon\u002Futil\u002FOrderedExecutor.html",[264],"OrderedExecutor"," works. BookKeeper provides OrderedExecutor in org.apache.bookkeeper.common.util. It guarantees that tasks with the same key are executed in the same thread. As we can see from the code snippet below, if we provide the same ordering key, we will always return the same thread with chooseThread. It helps us keep the order of tasks. When sending messages, Pulsar can run sustaining jobs with the same key, ensuring messages are sent in the expected order. This is widely used in Pulsar.",[1462,1463,1468],"pre",{"className":1464,"code":1466,"language":1467},[1465],"language-text","public void executeOrdered(Object orderingKey, Runnable r) {\n        chooseThread(orderingKey).execute(r);\n}\n","text",[1469,1470,1466],"code",{"__ignoreMap":18},[48,1472,1473],{},"We found two problems caused by OrderedExecutor according to our test results.",[48,1475,1476],{},"First, when we split 100K consumers into different Broadcast Subscriptions, the latency did not change too much. For example, we created four Broadcast Subscriptions with 25K consumers attached to each of them and hoped this approach would further reduce latency given its parallelization. In addition, dividing consumers into different groups should also help the broker have better communication with BookKeeper. However, we found that it had no noticeable effect on our benchmark results.",[48,1478,1479],{},"The reason is that Pulsar uses the topic name as the ordering key. This means that all the messages of the same tasks are sequentialized at the topic level. However, we know that subscriptions are independent of each other. It is unnecessary to guarantee the order across all the subscriptions. We just need to keep the message order within one subscription. A natural solution is to change the key to the subscription name.",[48,1481,1482],{},"The second one is more interesting. In terms of message acknowledgments, we noticed a very high long-tail latency. On average, acknowledgments finished in 0.5 seconds, but the slowest one took up to 7 seconds, which greatly affected the overall P99 latency. We carried out further research but did not find any problems in the network or consumers. This high latency issue could always be reproduced in every benchmark test.",[48,1484,1485],{},"Finally, we found that this issue was caused by the way Pulsar handles acknowledgments. Pulsar uses two individual tasks to complete the message-sending process - one for sending the message and the other for the ACK. For each message sent by the consumer, Pulsar generates these two tasks and pushes them to OrderedExecutor.",[48,1487,1488],{},[351,1489],{"alt":1490,"src":1491},"Figure of two tasks in the same thread","\u002Fimgs\u002Fblogs\u002F63be1b196d2f836330dc03f5_NTT-blog-image11.png",[48,1493,1494],{},"To guarantee the order of messages, Pulsar always adds them to the same thread, which is suitable for many use cases. However, things are slightly different when you have 100K consumers. As shown in Figure 12, Pulsar generates 200K tasks, all of which are inserted into a single thread. This means other tasks might also exist between a pair of SEND and ACK tasks. In these cases, Pulsar first runs the in-between tasks before the ACK task can be processed, leading to a longer latency. In a worst-case scenario, there might be 10,000 in-between tasks.",[48,1496,1497],{},[351,1498],{"alt":1499,"src":1500},"Figure shows other tasks might exist between a pair of SEND and ACK tasks.","\u002Fimgs\u002Fblogs\u002F63be1b65fffd8d8bce75a41d_NTT-blog-image12.png",[48,1502,1503],{},"For our case, we only need to send messages in order while their ACK tasks can be placed anywhere. Therefore, to solve this problem, we used a random thread for ACK tasks instead of the same thread. As shown in Table 3, our final test with the updated logic of OrderExecutor shows some promising results.",[48,1505,1506],{},[351,1507],{"alt":1508,"src":1509},"Table showing test results of Broadcast Subscription with improved OrderedExecutor","\u002Fimgs\u002Fblogs\u002F63be1b9f62fe43442351b3fe_Screen-Shot-2023-01-10-at-6.14.42-PM.png",[48,1511,1512],{},"Compared with the previous test using the original OrderedExecutor logic, the P99 latency in this test for 100K consumers was about 4 times shorter and the connection time was reduced by half. The latest design also worked well for 30K consumers, the connection time of which was about 2.5 times faster.",[40,1514,931],{"id":930},[48,1516,1517],{},"Pulsar has a flexible design and its performance is already good enough for many use cases. However, when you need to handle special cases where a large number of consumers exist, it may be a good idea to implement your own subscription model. This will help improve Pulsar’s performance dramatically.",[48,1519,1520],{},"Additionally, using OrderedExecutor in the right way is also important to the overall performance. When you have a large number of SEND and ACK tasks that need to be processed in a short time, you may want to optimize the original logic given the additional in-between tasks.",[40,1522,1524],{"id":1523},"more-on-apache-pulsar","More on Apache Pulsar",[339,1526,1527,1535,1550,1559],{},[342,1528,1529,1530,1534],{},"Make an inquiry: Interested in a fully-managed Pulsar offering built by the original creators of Pulsar? ",[55,1531,1533],{"href":1532},"\u002Fcontact\u002F","Contact us"," now.",[342,1536,1537,1538,1543,1544,1549],{},"Pulsar Summit Europe 2023 is taking place virtually on May 23rd. Engage with the community by ",[55,1539,1542],{"href":1540,"rel":1541},"https:\u002F\u002Fsessionize.com\u002Fpulsar-virtual-summit-europe-2023\u002F",[264],"submitting a CFP"," or ",[55,1545,1548],{"href":1546,"rel":1547},"https:\u002F\u002F6585952.fs1.hubspotusercontent-na1.net\u002Fhubfs\u002F6585952\u002FSponsorship%20Prospectus%20Pulsar%20Virtual%20Summit%20Europe%202023.pdf",[264],"becoming a community sponsor"," (no fee required).",[342,1551,1552,1553,1558],{},"Learn the Pulsar Fundamentals: Sign up for ",[55,1554,1557],{"href":1555,"rel":1556},"https:\u002F\u002Fwww.academy.streamnative.io\u002F",[264],"StreamNative Academy",", developed by the original creators of Pulsar, and learn at your own pace with on-demand courses and hands-on labs.",[342,1560,1561,1562,1566],{},"Read the ",[55,1563,1565],{"href":1564},"\u002Fblog\u002Fapache-pulsar-vs-apache-kafka-2022-benchmark","2022 Pulsar vs. Kafka Benchmark Report"," for the latest performance comparison on maximum throughput, publish latency, and historical read rate.",[48,1568,1569],{},"‍",{"title":18,"searchDepth":19,"depth":19,"links":1571},[1572,1573,1574,1580,1587,1588],{"id":1095,"depth":19,"text":1096},{"id":1126,"depth":19,"text":1127},{"id":1173,"depth":19,"text":1174,"children":1575},[1576,1577,1578,1579],{"id":1221,"depth":279,"text":1222},{"id":1242,"depth":279,"text":1243},{"id":1258,"depth":279,"text":1259},{"id":1274,"depth":279,"text":1275},{"id":1322,"depth":19,"text":1323,"children":1581},[1582,1583,1584,1585,1586],{"id":1346,"depth":279,"text":1347},{"id":1382,"depth":279,"text":1383},{"id":1412,"depth":279,"text":1413},{"id":1431,"depth":279,"text":1432},{"id":1447,"depth":279,"text":1448},{"id":930,"depth":19,"text":931},{"id":1523,"depth":19,"text":1524},"Apache Pulsar","2023-01-10","Learn why NTT chose Apache Pulsar for smart city and how Pulsar broadcasts messages for 100K consumers with an end-to-end latency of less than 1 second.","\u002Fimgs\u002Fblogs\u002F63be1ce6b7a9a44b9a5eb1de_pulsar-ntt-one-topic-100k-top.jpg",{},"\u002Fblog\u002Fhandling-100k-consumers-with-one-pulsar-topic","14 min read",{"title":1088,"description":1591},"blog\u002Fhandling-100k-consumers-with-one-pulsar-topic",[1589],"7N9HL6kVvEGB5N-WZ7oFmxb6O2Tso693rVxGbXCrBMI",[1601],{"id":1602,"title":1090,"bioSummary":1603,"email":289,"extension":8,"image":1604,"linkedinUrl":289,"meta":1605,"position":1612,"stem":1613,"twitterUrl":289,"__hash__":1614},"authors\u002Fauthors\u002Fhongjie-zhai.md","Hongjie Zhai is a Research Engineer at the Software Innovation Center of Nippon Telegraph and Telephone (NTT). He focuses on the improvement of data platforms (message brokers, databases, etc.) toward the next-generation hardware and network. Previously, he worked on machine learning platforms and container technologies. In addition to his main job, he is also doing research on deep learning and machine learning algorithms.","\u002Fimgs\u002Fauthors\u002Fhongjie-zhai.jpeg",{"body":1606},{"type":15,"value":1607,"toc":1610},[1608],[48,1609,1603],{},{"title":18,"searchDepth":19,"depth":19,"links":1611},[],"Research Engineer, Nippon Telegraph and Telephone Software Innovation Center","authors\u002Fhongjie-zhai","wEQmLM7O6xnbOoLqhawODn2MxUzZWvBSHiJU55-mdno",[1616,1624,1631],{"path":1617,"title":1618,"date":1619,"image":1620,"link":-1,"collection":1621,"resourceType":1622,"score":1623,"id":1617},"\u002Fsuccess-stories\u002Fgetui","Build a Priority-based Push Notification System Using Apache Pulsar at GeTui","2022-12-27","\u002Fimgs\u002Fsuccess-stories\u002F67942ee3c017499ff6794b64_SN-SuccessStories-GeTui.webp","successStories","Case Study",1.1,{"path":1625,"title":1626,"date":1627,"image":1628,"link":-1,"collection":1629,"resourceType":1630,"score":1623,"id":1625},"\u002Fwebinars\u002Fhow-to-operate-pulsar-in-production","How to Operate Pulsar in Production","2022-12-23","\u002Fimgs\u002Fwebinars\u002F63a5a1346ac3009373b5c4c0_OG_webinar-How%20to%20Operate%20Pulsar%20in%20Production.webp","webinars","Webinar",{"path":1632,"title":1633,"date":1627,"image":1634,"link":-1,"collection":1629,"resourceType":1630,"score":1623,"id":1632},"\u002Fwebinars\u002Flessons-from-managing-a-pulsar-cluster-by-nutanix","Lessons From Managing A Pulsar Cluster by Nutanix","\u002Fimgs\u002Fwebinars\u002F63a59f3960075d23d546447c_OG_webinar-Lessons%20From%20Managing%20A%20Pulsar%20Cluster%20by%20Nutanix.webp",1776256529177]