Taking Charge of Tables: Introducing OpenHouse for Big Data Management
Co-Authors: Sumedh Sakdeo, Lei Sun, Sushant Raikar, Stanislav Pak, and Abhishek Nath
Introduction
At LinkedIn, we build and operate an open source data lakehouse deployment to power Analytics and Machine Learning workloads. Leveraging data to drive decisions allows us to serve our members with better job insights, and connect the world’s professionals with each other.
Open source data lakehouse deployments are built on the foundations of compute engines (like Apache Spark, Trino, Apache Flink), distributed storage (HDFS, cloud blob stores), and metadata catalogs / table formats (like Apache Iceberg, Delta, Hudi, Apache Hive Metastore). End-users create relational entities in the form of Tables over structured or semi-structured data using compute engines, with the metadata for a Table stored in a catalog, and data stored in distributed storage.
While functional, our current setup for managing tables is fragmented. The individual building blocks of compute engines, distributed storage, and metadata catalogs operate independently as part of an overall data plane. Unfortunately, there is currently no system in open source that unifies them through a single control plane. This unification is crucial for simplifying lakehouse management, organizing data for optimal query performance, instituting governance, and declarative metadata management – all to provide an enhanced developer experience. As a result, data scientists, data engineers, and product engineers have to juggle multiple systems and manage tables individually. It adds toil in terms of complexity and potential inconsistencies that can serve as distractions to the developers’ core product focus. What developers are asking for is a way to declaratively specify the table definitions and policies using an API such as SQL, and the lakehouse should take care of the rest.
To provide an experience designed to reduce toil for product engineering and take charge of tables, we built and deployed OpenHouse, a control plane that allows our developers to interface with managed tables in our open source data lakehouse.
In this blog post, we will discuss the guiding principles outlined for OpenHouse and the northstar UX when interfacing with OpenHouse tables. We’ll also introduce OpenHouse’s control plane, specifics of the deployed system at LinkedIn including our managed Iceberg lakehouse, and the impact and roadmap for future development of OpenHouse, including a path to open source.
OpenHouse for Big Data Management
When building OpenHouse, we followed these four guiding principles to ensure that data platform teams and big data users could self-serve the creation of fully managed, publicly shareable, and governed tables in open source lakehouse deployments.
-
Tables (not files/blobs) are the only API abstraction for end-users. All accesses to table data must go through a table interface; no direct read/write is permitted to files or blobs on distributed storage for tabular data.
-
Tables are stored in a protected storage namespace that the control plane has full control over. Having full control allows the control plane to be opinionated about management aspects such as data organization, transactional semantics, security, high availability, disaster recovery, and quotas.
-
Tables are governed as per agreed upon company standards. This allows organizations to enforce constraints on data models, compliance annotations, and other metadata.
-
Tables are maintained regularly. This includes optimizing performance by adjusting sorting, partitioning, clustering strategies based on query statistics, and finally garbage collecting expired versions.
Figure 1: Northstar UX
Figure 1 shows the northstar user experience OpenHouse is building towards. This flow allows users to create a table, manipulate table metadata, load data, and share the table with a single chain of API calls, without losing their train of thought. In this user experience, most of the API calls can be made by leveraging standard SQL or Dataframe syntax.
-- create table in openhouse CREATE TABLE openhouse.db.table (id bigint COMMENT 'unique id', data string); -- manipulate table metadata ALTER TABLE openhouse.db.table_partitioned SET POLICY ( RETENTION=30d ); ALTER TABLE openhouse.db.table ALTER COLUMN measurement TYPE double; ALTER TABLE openhouse.db.table SET TBLPROPERTIES ('key1' = 'value1'); -- manipulate table data INSERT INTO openhouse.db.table VALUES ('1', 'a'); -- share table ALTER TABLE openhouse.db.table_partitioned SET POLICY ( SHARING=true ); GRANT SELECT ON TABLE openhouse.db.table TO user;Control Plane for Tables
The core of OpenHouse’s control plane is a RESTful Table Service that provides secure and scalable table provisioning and declarative metadata management. Furthermore, it can be configured to automatically orchestrate data services that keep the tables in user configured (e.g., retention, replication), optimal (e.g., storage compaction, sorting, clustering) and compliant state (e.g., GDPR purge). Figure 2 shows how OpenHouse fits into broader open source lakehouse deployments.
Figure 2: OpenHouse Control Plane
Table service acts as a central metadata repository (i.e., a catalog). At its core, table service exposes standard catalog APIs that allow users to perform CRUD operations on managed OpenHouse tables. In many ways, this can be seen as an evolution of Hive metastore, with these additional capabilities:
-
Table service offers declarative table management APIs, i.e., a client only needs to provide the desired state for a managed table. The table service works with data services to guarantee that the observed state of the table is reconciled to the desired state.
-
Table service provides a way to securely share the tables, with built in role-based access control for table operations. Additionally, it abstracts away all the underlying FileSystem and BlobStore permissioning schemes from the end-user.
-
Table service acts as a gateway to enforce data quality constraints, governance rules, and data modeling standards.
-
Table service is opinionated about how tables are laid out in an HDFS namespace or Blob Store bucket and how quotas are managed.
-
Core table service APIs are designed to allow support for multiple table formats, specifically, Iceberg, Delta, and Hudi. Any format specific features are implemented as API extensions, without impacting the core Table APIs.
-
Table service is built to be horizontally scalable, prevents noisy neighbors, and provides granular observability into table access patterns.
Data services are a set of table maintenance jobs that keep the underlying storage in a healthy state. These include a wide variety of built-in compaction jobs that optimize table storage to reduce load on the data storage system and optimize user queries, purger jobs that keep the tables in a compliant state, and cross-cluster replication jobs for disaster recovery. The framework itself is extensible to run custom jobs.
Deployed system at LinkedIn
Figure 3: Deployed System
Figure 3 shows system components of OpenHouse deployed at LinkedIn. Each component is numbered and its purpose is as follows:
-
Table service: This is a RESTful web service that exposes tables REST resources. This service is deployed on a Kubernetes cluster with a fronting Envoy Network Proxy.
-
REST clients: A variety of applications use REST clients to call into table service (#1). Clients include but are not limited to compliance apps, replication apps, data discovery apps like Datahub and IaC, Terraform providers, and data quality checkers. Some of the apps that work on all the tables in OpenHouse are assigned higher privileges.
-
Metastore Catalog: Spark,Trino, andFlink engines are a special flavor of REST clients. An OpenHouse specific metastore catalog implementation allows engines to integrate with OpenHouse tables.
-
House database service: This is an internal service to store table service and data service metadata. This service exposes a key-value interface that is designed to use a NoSQL DB for scale and cost optimization. However the deployed system is currently backed by a MySQL instance, for ease of development and deployment.
-
Managed namespace: This is a managed HDFS namespace where tables are persisted in Iceberg table format. Table service is responsible for setting up the table directory structure with appropriate FileSystem permissioning. OpenHouse has a novel HDFS permissioning scheme that makes it possible for any ETL flow to publish directly to Iceberg tables and securely into a managed HDFS namespace.
-
Data services: This is a set of data services that reconciles the user / system declared configuration with the system observed configuration. This includes use cases such as retention, restatement, and Iceberg-specific maintenance. Each maintenance activity is scheduled as a Spark job per table. A Kubernetes cronjob is run periodically on a schedule to trigger a maintenance activity. All the bookkeeping of jobs is done in House Database Service using a jobs metadata table for ease of debugging and monitoring.
Architecturally, OpenHouse is built to run in any cloud environment, using blob stores, managed compute, and cloud databases. Both the table service and data service are packaged as containers that should make it easy to deploy in a diverse environment. We are working on Terraform recipes that would automate deployment of the entire stack in minutes.
Managed Iceberg Lakehouse
At LinkedIn, OpenHouse tables are persisted on HDFS in Iceberg table format. Compared to Hive table format, Iceberg allows us to improve the reliability of tables on HDFS by providing features like incremental data processing, snapshot isolation, ACID transactions, and reproducible data flows through time travel queries.
Building a functional, scalable and easy to use lakehouse architecture with Iceberg as the table format required us to make new foundational investments. We invested in various data services that can work with Iceberg table format.
-
To keep the tables optimal, we automated orchestration of Iceberg maintenance jobs such as snapshot expiration, orphan file deletion, quarantine zones for deleted files, and manifest compaction.
-
To keep tables compliant, we have built data services that can delete data based on user requested purging and time partition expiration.
-
To provide data disaster recovery, we have built a data service that can replicate Iceberg snapshots efficiently across data centers.
Finally all our data services can be triggered almost instantaneously as Iceberg snapshots are committed.
Impact
In LinkedIn’s data lakes, two distinct categories of tables have emerged: centrally managed tables and self-managed tables. Centrally managed tables offer public sharing capabilities and robust table management support, including compaction and replication. On the other hand, self-managed tables are private to end-users and lack consistent management practices. Surprisingly, 65% of tables fall under the self-managed category, indicating a need for a more streamlined approach.
Our central managed platform imposes a laborious onboarding process, burdened by human intervention, resulting in significant time investment. It takes 2 to 3 weeks to onboard tables, and the ingestion is eventually consistent, creating operational complexities for both Site Reliability Engineers (SREs) and end-users.
With OpenHouse, end-users can self-serve creation of centrally managed, publicly shareable, and compliant tables in seconds. By eliminating the friction and operational complexities of traditional onboarding processes, OpenHouse empowers end-users to collaborate effectively while ensuring granular table sharing and adherence to compliance requirements, thereby transforming the way data lakes are operated.
Roadmap
OpenHouse has been deployed since late 2022 and serves a portion of our production traffic from LinkedIn’s GoToMarket systems that support LinkedIn Sales and Marketing. Our data engineers and data scientists who use dBT to create ETL flows were among the first to utilize this new system. Over the coming quarters, we will ramp production to serve the entirety of LinkedIn’s data lakehouse tables. We expect to share more details in future posts as well as any further plans to open source this technology early next year.
Acknowledgements
Big thanks to team members who have relentlessly shipped incremental milestones and delivered customer impact for this multi-year initiative: Lei Sun (founding engineer), Sushant Raikar, Stanislav Pak, Abhishek Nath, Malini Venkatachari, Rohit Kumar, Levi Jiang, Manisha Kamal, Swathi Koundinya, Vishal Saxena, and Naveen Selvaraj.
Over a year and half ago, OpenHouse was incubated under the leadership of Sumitha Poornachandran and she remains our unwavering pillar of support. Also, huge thanks to continuous support to our executive leadership Renu Tewari, Kartik Paramasivam, and Raghu Hiremagalur for believing in OpenHouse.
Many thanks to the thought leadership of Eric Baldeschwieler, Owen O’ Malley, Sriram Rao, Vasanth Rajamani, and Kapil Surlaker, who helped shape the product value proposition. Also we are grateful to peer reviewers for the blog, Erik Krogen, Daniel Meredith, and Diego Buthay.
Finally, OpenHouse is a product of many passionate discussions with leads across LinkedIn: Walaa Eldin Moustafa, Bhupendra Jain, Ratandeep Ratti, Kip Kohn, Issac Buenros and Maneesh Varshney.
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
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
CCI Said to Have Appointed Former WhatsApp Executive, Government Officials as New Members
WhatsApp iPad Support Spotted in Testing on Latest iOS Beta, Improved Group Calls Interface on Android
Allowing Users to Promote Stories as Ads (via Marketing API)
Facebook Hashtags: Free Hashtag Tool + 70 Ideas
Enhancing homepage feed relevance by harnessing the power of large corpus sparse ID embeddings
![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 Security and Developer Productivity: LinkedIn’s Journey with Implementing Content Security Policy
-
Uncategorized1 week ago3 Ways To Find Your Instagram Reels History
-
Uncategorized2 weeks agoSocial Media for Business: A Practical Guide
-
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
-
LINKEDIN1 week agoCareer Stories: Learning and growing through mentorship and community
-
OTHER2 weeks agoYouTube ‘Subscribe’ Button Spotted to Be Glowing When Creators Request Subscription
-
Uncategorized1 week agoCommunity Manager: Job Description & Key Responsibilities
-
Uncategorized7 days agoThe Complete Guide to Social Media Video Specs in 2023