Open Sourcing iris-message-processor
One measure of a successful network is uptime – providing consistent, reliable service for members and customers. If there are frequent connection errors or downtime notifications, it becomes difficult to deliver an experience where people can connect and interact with ease. When faced with uptime challenges, being able to quickly escalate issues to network engineers helps ensure that people can work the way that they want to.
At LinkedIn, escalations encompass various events, including alerts, system change notifications, and automated actions that require an engineer’s acknowledgment to proceed. These events follow a customizable escalation plan that generates notifications (often with increasing urgency) until an engineer claims the event or the needed steps are completed.
To manage our on-call escalations, we built Iris and Oncall, two open-source tools that we introduced to the community approximately six years ago. Oncall enables our teams to efficiently handle their on-call shifts through automated scheduling and a suite of calendar management tools. Iris leverages the data provided by Oncall to promptly alert on-call engineers in case of any issues and escalate matters if required. Developers have the ability to create personalized escalation plans and message templates, granting them control over who receives alerts and the specific content delivered in those alerts. Because of its ease of use and flexibility, Iris has also become LinkedIn’s internal message delivery platform, sending out alerts, deployment notifications, security notices, and more.
Together, these tools deliver flexibility, customization, and simplicity in managing on-call escalations and can be used as low-cost replacements for off-the-shelf incident response platforms like PagerDuty. Currently, Iris and Oncall have more than 350 forks and 1,700 stars on Github.
In this post, we’ll discuss how we used Iris to both scale up (~2000%) and speed up (~86x) our incident management system. We’ll also share how this journey resulted in an incredibly robust and effective system that we open-sourced, allowing it to be easily and freely deployed by any company.
Growth of Iris within LinkedIn
When we first open sourced Iris, we published this blog post that included the following graph showing the month-by-month growth of Iris escalations at LinkedIn. At the time, adoption of Iris had steadily been growing and Iris was integrated into the major components of our alerting infrastructure.
Figure 1: Graph showing monthly Iris escalations 2015-2017
Six years later that graph looks very different. Iris has become a ubiquitous service at LinkedIn with a 700% increase in the number of other services directly integrated via its API. At the same time, the scope and complexity of LinkedIn’s engineering footprint has grown massively. As a result, Iris experienced a 2,300% growth in the number of escalations Iris now processes monthly.
Figure 2: Graph showing monthly Iris escalations in 2023
The difference is greater still when we look at messages sent by Iris. The role expansion of Iris as a generalized message sending API made its usage skyrocket. Iris currently sends, on average, over 700,000 messages daily with bursts of more than 3,000 messages per second!
Figure 3: Graph showing daily Iris messages June 2023
At this scale we were starting to run into the limits within Iris’ original design and we recognized that changes were needed to continue meeting our very strict reliability and performance standards.
Designing for scale
When it became clear that the existing design would not be able to continue serving our needs, we embarked on a project to re-architect Iris into a service that could scale across the next “10x” of growth at LinkedIn, minimizing the need for future redesigns. To do so, we first had to identify what exactly were some of the most pressing issues with the current design.
One of the most crucial issues that we wanted to address was the way Iris-api handles message processing and escalations. Previously, Iris-api relied on an iris-sender python subprocess running on a single leader node that would ingest all the escalations from the database. It would then evaluate and render each message one-by-one before passing it off to other senders in the cluster for sending. Because it processed escalations and rendered messages serially, any sizable enough burst of escalations could cause massive delays, up to tens of minutes in the worst cases. Failures of the leader also had the potential to cause outsized impact, if the sender process got stuck all escalation processing would completely grind to a halt.
Additionally, the iris-sender design relied on having strong consistency in its database which we achieved by using a Galera cluster. However, as the volume of escalations and especially messages grew dramatically, the demands started outpacing the capabilities of the system. The single leader iris-sender subprocess used the database as a message queue, which was becoming more untenable as the volume of messages grew – especially since both the Iris-api and the Galera clusters were all geographically distributed across three different data centers. This resulted in several issues, mainly the occurrence of Galera replication deadlocks that caused requests to fail intermittently.
To address all these issues we created a new service written in Go called iris-message-processor. iris-message-processor is a fully distributed replacement for the iris-sender subprocess. With this new service, Iris escalations are split up into buckets which are dynamically assigned/reassigned to different iris-message-processor nodes as they join or leave the cluster. In turn, these iris-message-processor nodes concurrently process their escalations and messages assigned to them as well as directly send them out to the appropriate vendors. This means that instead of relying on a single sender leader, the iris-message-processor cluster can now be horizontally scaled with virtually no limits to accommodate ever expanding escalation or message volume. Additionally, because the database is no longer used as a message queue, the demands on the existing cluster are much lower and a separate easier to scale eventually consistent database can be used to store the resulting messages for tracking.
An additional bonus of doing a major rewrite is that it gave us the opportunity to tackle some other long standing tech debt and add features to improve reliability and performance. These include a global per mode rate limit to prevent rate limiting by vendors like Slack; Per application per mode round robin distribution of message rate limits so that we could accept large concurrent volumes of messages from a specific client without degrading another client’s experience; Prioritization of critical alerting escalations over “out-of-band notifications”; Better introspection into the escalation and message queue, and more.
Figure 4. Architecture diagram of the new Iris ecosystem
Performance comparison
The performance gains we were able to achieve with the iris-message-processor were a great success. To test the effectiveness of our changes we load tested Iris escalation processing before and after switching to Iris-message processor and measured the time between an escalation creation request being received and the delivery of that escalation’s messages to the appropriate vendors. We found that:
-
Under an average load (100 escalations per minute) iris-message-processor was ~4.6x faster than iris-sender
-
Under a high load (2,000 escalations per minute) iris-message-processor was ~86.6x faster than iris-sender
We also recreated a previous outage scenario where a burst of 6,000 simultaneous escalations created a massive slowdown in Iris escalation processing that took almost 30 minutes to clear as shown in Figure 5:
Figure 5: Graphs detailing old iris-sender processing ~6,000 escalation burst
When submitted to the same load, the iris-message-processor took less than 10 seconds to process all the escalations.
Figure 6: Graphs detailing new iris-message-processor processing ~6,000 escalation burst
We even tested stopping 50% of the iris-message-processor nodes simultaneously to test the automatic rebalancing of escalation buckets to great success. The whole cluster automatically rebalanced in less than 30 seconds and escalation time (the time it takes to process all currently active escalations) stayed under three seconds even under an above average load.
Figure 7: Graph detailing iris-message-processor rebalancing its nodes after dropping 50% of member nodes
A system can be tested very thoroughly but what ultimately matters is how it performs in a real world production scenario. At the time of writing we have now been running on iris-message-processor at LinkedIn for about a year with no outages and consistently beating SLOs of 1000ms/msg. In that time we have encountered many cases which would have caused issues with the previous iris-sender but so far the iris-message-processor has proven to be a worthwhile investment.
How to use iris-message-processor
Iris was and continues to be one of the many critical pieces of infrastructure at LinkedIn. Because of that, we wanted to make our changes as transparent as possible to anyone using or interfacing with Iris. We achieved that by keeping the existing Iris-api as is and having an iris-message-processor just act as a drop-in replacement for the iris-sender subprocess. By doing so we can maintain the same UI and API interfaces while delivering performance gains under the hood.
Additionally we took extra measures to ensure the stability of the platform during the rollout. This included preserving backwards compatibility with the iris-sender script so it was easy to toggle back to using iris-sender in case of an outage as well as designing the system with the capability to gradually ramp a percentage of messages and escalations for verification and testing purposes. At this point the iris-message processor has been thoroughly battle tested so it is safe to fully move to processing all messages and escalations with the iris-message-processor directly.
The iris-message-processor can be installed and run separately and through just a few simple configuration changes on the Iris-api side they will seamlessly talk to each other through the existing Iris-api. Full instructions and code can be found in the iris-message-processor repo.
We recently released iris-message-processor as open source software here to join the existing Iris and Oncall repos. These projects are meant to be able to operate outside of LinkedIn’s internal environment and serve as complete replacements for other existing off the shelf incident management systems within any environment.
We welcome any potential users or contributors. You can also reach our team with general questions about the project by opening an issue in the GitHub repository or emailing iris-oncall@linkedin.com.
Acknowledgements
The iris-message-processor project was made possible through the efforts of Diego Cepeda, Joe Gillotti, James Won, Colin Yang, and Fellyn Silliman. Additional thanks to Sam Moffatt, Kahnan Patel, and Michael Herstine for their invaluable support.
Topics
Since learning frontend and backend skills, Rishika’s passion for engineering has expanded beyond her team at LinkedIn to grow into her own digital community. As she develops as an engineer, giving back has become the most rewarding part of her role.
From intern to engineer—life at LinkedIn
My career with LinkedIn began with a college internship, where I got to dive into all things engineering. Even as a summer intern, I absorbed so much about frontend and backend engineering during my time here. When I considered joining LinkedIn full-time after graduation, I thought back to the work culture and how my manager treated me during my internship. Although I had a virtual experience during COVID-19, the LinkedIn team ensured I was involved in team meetings and discussions. That mentorship opportunity ultimately led me to accept an offer from LinkedIn over other offers.
Before joining LinkedIn full-time, I worked with Adobe as a Product Intern for six months, where my projects revolved around the core libraries in the C++ language. When I started my role here, I had to shift to using a different tech stack: Java for the backend and JavaScript framework for the frontend. This was a new challenge for me, but the learning curve was beneficial since I got hands-on exposure to pick up new things by myself. Also, I have had the chance to work with some of the finest engineers; learning from the people around me has been such a fulfilling experience. I would like to thank Sandeep and Yash for their constant support throughout my journey and for mentoring me since the very beginning of my journey with LinkedIn.
Currently, I’m working with the Trust team on building moderation tools for all our LinkedIn content while guaranteeing that we remove spam on our platform, which can negatively affect the LinkedIn member experience. Depending on the project, I work on both the backend and the frontend, since my team handles the full-stack development. At LinkedIn, I have had the opportunity to work on a diverse set of projects and handle them from end to end.
Mentoring the next generation of engineering graduates
I didn’t have a mentor during college, so I’m so passionate about helping college juniors find their way in engineering. When I first started out, I came from a biology background, so I was not aware of programming languages and how to translate them into building a technical resume. I wish there would have been someone to help me out with debugging and finding solutions, so it’s important to me to give back in that way.
I’m quite active in university communities, participating in student-led tech events like hackathons to help them get into tech and secure their first job in the industry. I also love virtual events like X (formally Twitter) and LinkedIn Live events. Additionally, I’m part of LinkedIn’s CoachIn Program, where we help with resume building and offer scholarships for women in tech.
Influencing online and off at LinkedIn
I love creating engineering content on LinkedIn, X, and other social media platforms, where people often contact me about opportunities at LinkedIn Engineering. It brings me so much satisfaction to tell others about our amazing company culture and connect with future grads.
When I embarked on my role during COVID-19, building an online presence helped me stay connected with what’s happening in the tech world. I began posting on X first, and once that community grew, I launched my YouTube channel to share beginner-level content on data structures and algorithms. My managers and peers at LinkedIn were so supportive, so I broadened my content to cover aspects like soft skills, student hackathons, resume building, and more. While this is in addition to my regular engineering duties, I truly enjoy sharing my insights with my audience of 60,000+ followers. And the enthusiasm from my team inspires me to keep going! I’m excited to see what the future holds for me at LinkedIn as an engineer and a resource for my community on the LinkedIn platform.
About Rishika
Rishika holds a Bachelor of Technology from Indira Gandhi Delhi Technical University for Women. Before joining LinkedIn, she interned at Google as part of the SPS program and as a Product Intern at Adobe. She currently works as a software engineer on LinkedIn’s Trust Team. Outside of work, Rishika loves to travel all over India and create digital art.
Editor’s note: Considering an engineering/tech career at LinkedIn? In this Career Stories series, you’ll hear first-hand from our engineers and technologists about real life at LinkedIn — including our meaningful work, collaborative culture, and transformational growth. For more on tech careers at LinkedIn, visit: lnkd.in/EngCareers.
Lekshmy has always been interested in a role in a company that would allow her to use her people skills and engineering background to help others. Working as a software engineer at various companies led her to hear about the company culture at LinkedIn. After some focused networking, Lekshmy landed her position at LinkedIn and has been continuing to excel ever since.
How did I get my job at LinkedIn? Through LinkedIn.
Before my current role, I had heard great things about the company and its culture. After hearing about InDays (Investment Days) and how LinkedIn supports its employees, I knew I wanted to work there.
While at the College of Engineering, Trivandrum (CET), I knew I wanted to pursue a career in software engineering. Engineering is something that I’m good at and absolutely love, and my passion for the field has only grown since joining LinkedIn. When I graduated from CET, I began working at Groupon as a software developer, starting on databases, REST APIs, application deployment, and data structures. From that role, I was able to advance into the position of software developer engineer 2, which enabled me to dive into other software languages, as well as the development of internal systems. That’s where I first began mentoring teammates and realized I loved teaching and helping others. It was around this time that I heard of LinkedIn through the grapevine.
Joining the LinkedIn community
Everything I heard about LinkedIn made me very interested in career opportunities there, but I didn’t have connections yet. I did some research and reached out to a talent acquisition manager on LinkedIn and created a connection which started a path to my first role at the company.
When I joined LinkedIn, I started on the LinkedIn Talent Solutions (LTS) team. It was a phenomenal way to start because not only did I enjoy the work, but the experience served as a proper introduction to the culture at LinkedIn. I started during the pandemic, which meant remote working, and eventually, as the world situation improved, we went hybrid. This is a great system for me; I have a wonderful blend of being in the office and working remotely. When I’m in the office, I like to catch up with my team by talking about movies or playing games, going beyond work topics, and getting to know each other. With LinkedIn’s culture, you really feel that sense of belonging and recognize that this is an environment where you can build lasting connections.
LinkedIn: a people-first company
If you haven’t been able to tell already, even though I mostly work with software, I truly am a people person. I just love being part of a community. At the height of the pandemic, I’ll admit I struggled with a bit of imposter syndrome and anxiety. But I wasn’t sure how to ask for help. I talked with my mentor at LinkedIn, and they recommended I use the Employee Assistance Program (EAP) that LinkedIn provides.
I was nervous about taking advantage of the program, but I am so happy that I did. The EAP helped me immensely when everything felt uncertain, and I truly felt that the company was on my side, giving me the space and resources to help relieve my stress. Now, when a colleague struggles with something similar, I recommend they consider the EAP, knowing firsthand how effective it is.
Building a path for others’ growth
With my mentor, I was also able to learn about and become a part of our Women in Technology (WIT) WIT Invest Program. WIT Invest is a program that provides opportunities like networking, mentorship check-ins, and executive coaching sessions. WIT Invest helped me adopt a daily growth mindset and find my own path as a mentor for college students. When mentoring, I aim to build trust and be open, allowing an authentic connection to form. The students I work with come to me for all kinds of guidance; it’s just one way I give back to the next generation and the wider LinkedIn community. Providing the kind of support my mentor gave me early on was a full-circle moment for me.
Working at LinkedIn is everything I thought it would be and more. I honestly wake up excited to work every day. In my three years here, I have learned so much, met new people, and engaged with new ideas, all of which have advanced my career and helped me support the professional development of my peers. I am so happy I took a leap of faith and messaged that talent acquisition manager on LinkedIn. To anyone thinking about applying to LinkedIn, go for it. Apply, send a message, and network—you never know what one connection can bring!
About Lekshmy
Based in Bengaluru, Karnataka, India, Lekshmy is a Senior Software Engineer on LinkedIn’s Hiring Platform Engineering team, focused on the Internal Mobility Project. Before joining LinkedIn, Lekshmy held various software engineering positions at Groupon and SDE 3. Lekshmy holds a degree in Computer Science from the College of Engineering, Trivandrum, and is a trained classical dancer. Outside of work, Lekshmy enjoys painting, gardening, and trying new hobbies that pique her interest.
Editor’s note: Considering an engineering/tech career at LinkedIn? In this Career Stories series, you’ll hear first-hand from our engineers and technologists about real life at LinkedIn — including our meaningful work, collaborative culture, and transformational growth. For more on tech careers at LinkedIn, visit: lnkd.in/EngCareers.
Topics
Espresso is the database that we designed to power our member profiles, feed, recommendations, and hundreds of other Linkedin applications that handle large amounts of data and need both high performance and reliability. As Espresso continued to expand in support of our 950M+ member base, the number of network connections that it needed began to drive scalability and resiliency challenges. To address these challenges, we migrated to HTTP/2. With the initial Netty based implementation, we observed a 45% degradation in throughput which we needed to analyze then correct.
In this post, we will explain how we solved these challenges and improved system performance. We will also delve into the various optimization efforts we employed on Espresso’s online operation section, implementing one approach that resulted in a 75% performance boost.
Espresso Architecture
Figure 1. Espresso System Overview
Figure 1 is a high-level overview of the Espresso ecosystem, which includes the online operation section of Espresso (the main focus of this blog post). This section comprises two major components – the router and the storage node. The router is responsible for directing the request to the relevant storage node and the storage layer’s primary responsibility is to get data from the MySQL database and present the response in the desired format to the member. Espresso utilizes the open-source framework Netty for the transport layer, which has been heavily customized for Espresso’s needs.
Need for new transport layer architecture
In the communication between the router and storage layer, our earlier approach involved utilizing HTTP/1.1, a protocol extensively employed for interactions between web servers and clients. However, HTTP/1.1 operates on a connection-per-request basis. In the context of large clusters, this approach led to millions of concurrent connections between the router and the storage nodes. This resulted in constraints on scalability, resiliency, and numerous performance-related hurdles.
Scalability: Scalability is a crucial aspect of any database system, and Espresso is no exception. In our recent cluster expansion, adding an additional 100 router nodes caused the memory usage to spike by around 2.5GB. The additional memory can be attributed to the new TCP network connections within the storage nodes. Consequently, we experienced a 15% latency increase due to an increase in garbage collection. The number of connections to storage nodes posed a significant challenge to scaling up the cluster, and we needed to address this to ensure seamless scalability.
Resiliency: In the event of network flaps and switch upgrades, the process of re-establishing thousands of connections from the router often breaches the connection limit on the storage node. This, in turn, causes errors and the router to fail to communicate with the storage nodes.
Performance: When using the HTTP/1.1 architecture, routers maintain a limited pool of connections to each storage node within the cluster. In some larger clusters, the wait time to acquire a connection can be as high as 15ms at the 95th percentile due to the limited pool. This delay can significantly affect the system’s response time.
We determined that all of the above limitations could be resolved by transitioning to HTTP/2, as it supports connection multiplexing and requires a significantly lower number of connections between the router and the storage node.
We explored various technologies for HTTP/2 implementation but due to the strong support from the open-source community and our familiarity with the framework, we went with Netty. When using Netty out of the box, the HTTP/2 implementation throughput was 45% less than the original (HTTP/1.1) implementation. Because the out of the box performance was very poor, we had to implement different optimizations to enhance performance.
The experiment was run on a production-like test cluster and the traffic is a combination of access patterns, which include read and write traffic. The results are as follows:
| Protocol | QPS | Single Read Latency (P99) | Multi-Read Latency (P99) |
| HTTP/1.1 | 9K | 7ms | 25ms |
| HTTP/2 | 5K (-45%) | 11ms (+57%) | 42ms (+68%) |
On the routing layer, after further analysis using flame graphs, major differences between the two protocols are shown in the following table.
| CPU overhead | HTTP/1.1 | HTTP/2 |
| Acquiring a connection and processing the request | 20% | 32% (+60%) |
| Encode/Decode HTTP request | 18% | 32% (+77%) |
Improvements to Request/Response Handling
Reusing the Stream Channel Pipeline
One of the core concepts of Netty is its ChannelPipeline. As seen in Figure 1, when the data is received from the socket, it is passed through the pipeline which processes the data. Channel Pipeline contains a list of Handlers, each working on a specific task.
Figure 2. Netty Pipeline
In the original HTTP/1.1 Netty pipeline, a set of 15-20 handlers was established when a connection was made, and this pipeline was reused for all subsequent requests served on the same connection.
However, in HTTP/2 Netty’s default implementation, a fresh pipeline is generated for each new stream or request. For instance, a multi-get request to a router with over 100 keys can often result in approximately 30 to 35 requests being sent to the storage node. Consequently, the router must initiate new pipelines for all 35 storage node requests. The process of creating and dismantling pipelines for each request involving a considerable number of handlers turned out to be notably resource-intensive in terms of memory utilization and garbage collection.
To address this concern, a forked version of Netty’s Http2MultiplexHandler has been developed to maintain a queue of local stream channels. As illustrated in Figure 2, on receiving a new request, the multiplex handler no longer generates a new pipeline. Instead, it retrieves a local channel from the queue and employs it to process the request. Subsequent to request completion, the channel is returned to the queue for future use. Through the reuse of existing channels, the creation and destruction of pipelines are minimized, leading to a reduction in memory strain and garbage collection.
Figure 3. Sequence diagram of stream channel reuse
Addressing uneven work distribution among Netty I/O threads
When a new connection is created, Netty assigns this connection to one of the 64 I/O threads. In Espresso, the number of I/O threads is equal to twice the number of cores present. The I/O thread associated with the connection is responsible for I/O and handling the request/response on the connection. Netty’s default implementation employs a rudimentary method for selecting an appropriate I/O thread out of the 64 available for a new channel. Our observation revealed that this approach leads to a significantly uneven distribution of workload among the I/O threads.
In a standard deployment, we observed that 20% of I/O threads were managing 50% of all the total connections/requests. To address this issue, we introduced a BalancedEventLoopGroup. This entity is designed to evenly distribute connections across all available worker threads. During channel registration, the BalancedEventLoopGroup iterates through the worker threads to ensure a more equitable allocation of workload
After this change, during registering of a channel, an event loop with the number of connections below the average is selected.
private EventLoop selectLoop() { int average = averageChannelsPerEventLoop(); EventLoop loop = next(); if (_eventLoopCount > 1 && isUnbalanced(loop, average)) { ArrayList list = new ArrayList<>(_eventLoopCount); _eventLoopGroup.forEach(eventExecutor -> list.add((EventLoop) eventExecutor)); Collections.shuffle(list, ThreadLocalRandom.current()); Iterator it = list.iterator(); do { loop = it.next(); } while (it.hasNext() && isUnbalanced(loop, average)); } return loop; } Reducing context switches when acquiring a connection
In the HTTP/2 implementation, each router maintains 10 connections to every storage node. These connections serve as communication pathways for the router I/O threads interfacing with the storage node. Previously, we utilized Netty’s FixedChannelPool implementation to oversee connection pools, handling tasks like acquiring, releasing, and establishing new connections.
However, the underlying queue within Netty’s implementation is not inherently thread-safe. To obtain a connection from the pool, the requesting worker thread must engage the I/O worker overseeing the pool. This process led to two context switches. To resolve this, we developed a derivative of the Netty pool implementation that employs a high-performance, thread-safe queue. Now, the task is executed by the requesting thread instead of a distinct I/O thread, effectively eliminating the need for context switches.
Improvements to SSL Performance
The following section describes various optimizations to improve the SSL performance.
Offloading DNS lookup and handshake to separate thread pool
During an SSL handshake, the DNS lookup procedure for resolving a hostname to an IP address functions as a blocking operation. Consequently, the I/O thread responsible for executing the handshake might be held up for the entirety of the DNS lookup process. This delay can result in request timeouts and other issues, especially when managing a substantial influx of incoming connections concurrently.
To tackle this concern, we developed an SSL initializer that conducts the DNS lookup on a different thread prior to initiating the handshake. This method involves passing the InetAddress, that contains both the IP address and hostname, to the SSL handshake procedure, effectively circumventing the need for a DNS lookup during the handshake.
Enabling Native SSL encryption/decryption
Java’s default built-in SSL implementation carries a significant performance overhead. Netty offers a JNI-based SSL engine that demonstrates exceptional efficiency in both CPU and memory utilization. Upon enabling OpenSSL within the storage layer, we observed a notable 10% reduction in latency. (The router layer already utilizes OpenSSL.)
To employ Netty Native SSL, one must include the pertinent Netty Native dependencies, as it interfaces with OpenSSL through the JNI (Java Native Interface). For more detailed information, please refer to https://netty.io/wiki/forked-tomcat-native.html.
Improvements to Encode/Decode performance
This section focuses on the performance improvements we made when converting bytes to Http objects and vice versa. Approximately 20% of our CPU cycles are spent on encode/decode bytes. Unlike a typical service, Espresso has very rich headers. Our HTTP/2 implementation involves wrapping the existing HTTP/1.1 pipeline with HTTP/2 functionality. While the HTTP/2 layer handles network communication, the core business logic resides within the HTTP/1.1 layer. Due to this, each incoming request required the conversion of HTTP/2 requests to HTTP/1.1 and vice versa, which resulted in high CPU usage, memory consumption, and garbage creation.
To improve performance, we have implemented a custom codec designed for efficient handling of HTTP headers. We introduced a new type of request class named Http1Request. This class effectively encapsulates an HTTP/2 request as an HTTP/1.1 by utilizing wrapped Http2 headers. The primary objective behind this approach is to avoid the expensive task of converting HTTP/1.1 headers to HTTP/2 and vice versa.
For example:
public class Http1Headers extends HttpHeaders { private final Http2Headers _headers; …. } And Operations such as get, set, and contains operate on the Http2Headers:
@Override public String get(String name) { return str(_headers.get(AsciiString.cached(name).toLowerCase()); } To make this possible, we developed a new codec that is essentially a clone of Netty’s Http2StreamFrameToHttpObjectCodec. This codec is designed to translate HTTP/2 StreamFrames to HTTP/1.1 requests/responses with minimal overhead. By using this new codec, we were able to significantly improve the performance of encode/decode operations and reduce the amount of garbage generated during the conversions.
Disabling HPACK Header Compression
HTTP/2 introduced a new header compression algorithm known as HPACK. It works by maintaining an index list or dictionaries on both the client and server. Instead of transmitting the complete string value, HPACK sends the associated index (integer) when transmitting a header. HPACK encompasses two key components:
-
Static Table – A dictionary comprising 61 commonly used headers.
-
Dynamic Table – This table retains the user-generated header information.
The Hpack header compression is tailored to scenarios where header contents remain relatively constant. But Espresso has very rich headers with stateful information such as timestamps, SCN, and so on. Unfortunately, HPACK didn’t align well with Espresso’s requirements.
Upon examining flame graphs, we observed a substantial stack dedicated to encoding/decoding dynamic tables. Consequently, we opted to disable dynamic header compression, leading to an approximate 3% enhancement in performance.
In Netty, this can be disabled using the following:
Http2FrameCodecBuilder.forClient() .initialSettings(Http2Settings.defaultSettings().headerTableSize(0));Results
Latency Improvements
| P99.9 Latency | HTTP/1.1 | HTTP/2 |
| Single Key Get | 20ms | 7ms (-66%) |
| Multi Key Get | 80ms | 20ms (-75%) |
We observed a 75% reduction in 99th and 99.9th percentile multi-read and read latencies, decreasing from 80ms to 20ms.
Figure 4. Latency reduction after HTTP/2
We observed similar latency reductions across the 90th percentile and higher.
Reduction in TCP connections
| HTTP/1.1 | HTTP/2 | |
| No of TCP Connections | 32 million | 3.9 million (-88%) |
We observed an 88% reduction in the number of connections required between routers and storage nodes in some of our largest clusters.
Figure 5. Total number of connections after HTTP/2
Reduction in Garbage Collection time
We observed a 75% reduction in garbage collection times for both young and old gen.
| GC | HTTP/1.1 | HTTP/2 |
| Young Gen | 2000 ms | 500ms (+75%) |
| Old Gen | 80 ms | 15 ms (+81%) |
Figure 6. Reduction in time for GC after HTTP/2
Waiting time to acquire a Storage Node connection
HTTP/2 eliminates the need to wait for a storage node connection by enabling multiplexing on a single TCP connection, which is a significant factor in reducing latency compared to HTTP/1.1.
| HTTP/1.1 | HTTP/2 | |
| Wait time in router to get a storage node connection | 11ms | 0.02ms (+99%) |
Figure 7. Reduction is wait time to get a connection after HTTP/2
Conclusion
Espresso has a large server fleet and is mission-critical to a number of LinkedIn applications. With HTTP/2 migration, we successfully solved Espresso’s scalability problems due to the huge number of TCP connections required between the router and the storage nodes. The new architecture also reduced the latencies by 75% and made Espresso more resilient.
Acknowledgments
I would like to thank my colleagues Antony Curtis, Yaoming Zhan, BinBing Hou, Wenqing Ding, Andy Mao, and Rahul Mehrotra who worked on this project. The project demanded a great deal of time and effort due to the complexity involved in optimizing the performance. I would like to thank Kamlakar Singh and Yun Sun for reviewing the blog and providing valuable feedback.
We would also like to thank our management Madhur Badal, Alok Dhariwal and Gayatri Penumetsa for their support and resources, which played a crucial role in the success of this project. Their encouragement and guidance helped the team overcome challenges and deliver the project on time.
Topics
How To Save Tons of Time With Instagram Auto Reply
Instagram Emoji Guide: Meanings, Reactions, Ideas
YouTube Announces AI-Enabled Editing Products for Video Creators
WhatsApp Passkey Support Reportedly Rolling Out to Beta Testers on Android: How It Works
Career stories: Influencing engineering growth at LinkedIn
Meta Urged Not to Roll Out End-to-end Encryption on Messenger, Instagram by UK
Allowing Users to Promote Stories as Ads (via Marketing API)
Facebook Hashtags: Free Hashtag Tool + 70 Ideas
![17-tips-for-writing-effective-youtube-descriptions-[free-tool]](https://www.social-ping.com/wp-content/uploads/2023/08/127190-17-tips-for-writing-effective-youtube-descriptions-free-tool-80x80.png)
17 Tips for Writing Effective YouTube Descriptions [Free Tool]
How to Boost Engagement With LinkedIn Carousel Posts and Ads
Enhancing homepage feed relevance by harnessing the power of large corpus sparse ID embeddings
Enhancing Security and Developer Productivity: LinkedIn’s Journey with Implementing Content Security Policy
-
Uncategorized1 week ago3 Ways To Find Your Instagram Reels History
-
OTHER2 weeks agoWhatsApp Chat Interoperability Feature Spotted in Development on Latest Beta Update: Report
-
FACEBOOK1 week agoIntroducing Facebook Graph API v18.0 and Marketing API v18.0
-
Uncategorized1 week agoCommunity Manager: Job Description & Key Responsibilities
-
OTHER2 weeks agoYouTube ‘Subscribe’ Button Spotted to Be Glowing When Creators Request Subscription
-
Uncategorized1 week agoThe Complete Guide to Social Media Video Specs in 2023
-
LINKEDIN1 week agoCareer Stories: Learning and growing through mentorship and community
-
Uncategorized1 week agoSocial Media Intelligence: What It Is & Why You Need It