P99 CONF is a new, cross-industry virtual event for _engineers_ and by engineers. The event is centered around low-latency, high-performance design — ranging from OS (kernel, eBPF, IO_uring), CPUs (Arm, Intel, OpenRisc), middleware and languages (go, Rust, JVM, DPDK), databases and observability methods.
Leading engineers will share their novel approaches for solving complex problems. The goal is to share development learnings, thus it is vendor and tool agnostic. It is not about products but about technology, so open source solutions are preferred. P99 CONF is for a highly technical audience only. Your boss is not invited.
Discover the latest methods in systems development and operational best practices for high-performance computing.
Share Your Team’s Ingenuity
Showcase your team’s success at achieving massive scale while maintaining lowest latencies. Compare notes with your industry peers.
Join the Webscale™ Revolution
Take lessons learned back to your organization and be part of the movement for ever-faster computing and big data solutions.
Henrik is a Cloud Native Advocate at Dynatrace, the leading Observability platform. Prior to Dynatrace, Henrik has worked as a Partner Solution Evangelist at Neotys, delivering webinars, building protypes to enhance the capability of NeoLoad. He has been working in the performance world more than 15 years, delivering projects in all contexts including extremely large Cloud testing on the most demanding business areas such as trading applications, Video on Demand, sports websites, etc. Henrik Rexed Is Also one of the Organizer of the Conference named Performance Advisory Council.
G1: To Infinity and Beyond
Stefan is a software engineer working in the HotSpot GC-team at Oracle. He joined the GC-team in 2013 and has been active in OpenJDK since then with a focus on the G1 garbage collector. Prior to working with GC in the JVM he was working on other JVM related projects at Oracle/Bea.
Understanding Apache Kafka P99 Latency at Scale
Pere is a Solution Architect for Confluent out of Berlin, Germany. He has been working with data and architecting systems for more than 15 years as a freelance engineer and consultant. In that role he was focused on data processing and search, helping companies build reliable and scalable data architectures. His work sits usually at the crossroad of infrastructure, data engineers and scientists, ontologist and product. Prior to that he was part of Elastic, the company behind Elasticsearch, where he was part of the Logstash team, helping companies build reliable ingestion pipelines into Elasticsearch. When not working, Pere loves to spend time with his lovely wife and kids, build Legos and enjoy Handball.
Scaling Apache Pulsar to 10 Petabytes/Day
Karthik Ramasamy is a Senior Director of Engineering at Splunk. Before Splunk, he was co-founder and CEO of Streamlio where he focused on building the next generation real time processing engine based on Apache Pulsar. He previously served as the engineering manager and technical lead for real-time analytics at Twitter where he co-created Twitter Heron. He has two decades of experience working in parallel databases, big data infrastructure, and networking. Karthik is the author of several publications, patents, and "Network Routing: Algorithms, Protocols and Architectures". He has a Ph.D. in computer science from the University of Wisconsin, Madison with a focus on big data and databases.
Whoops! I Rewrote It in Rust
Brian is a software engineer at Twitter where he focuses on performance and optimization projects. He contributes to several Twitter open source projects including: Pelikan, Rezolus, and rpc-perf. Brian is passionate about high performance software, systems tuning, monitoring, benchmarking, and Rust. When not hacking on code, Brian volunteers in his local search and rescue unit and is training his dog to find people who get lost in the wilderness.
Object Compaction in Cloud for High Yield
Tejas Chopra is a Senior Software Engineer, working in the Data Storage Platform team at Netflix, where he is responsible for architecting storage solutions to support exabytes of storage generated by Netflix Studios and Netflix Streaming Platform. Tejas has worked on distributed file systems & backend architectures, both in on-premise and cloud environments as part of several startups in his career & has a Masters Degree in Electrical & Computer Engineering from Carnegie Mellon University
Welcome to P99 CONF
Dor Laor is the CEO of ScyllaDB. Previously, Dor was part of the founding team of the KVM hypervisor under Qumranet that was acquired by Red Hat. At Red Hat Dor was managing the KVM and Xen development for several years. Dor holds an MSc from the Technion and a Phd in snowboarding.
Rust, Wright’s Law, and the Future of Low-Latency Systems
Bryan Cantrill is a software engineer who has spent a quarter of a century at the hardware/software interface. He is the co-founder and CTO of Oxide Computer Company, which is endeavoring to build a rack-scale computer for the post-cloud era. Prior to Oxide he spent nearly a decade at Joyent, a cloud computing pioneer; prior to Joyent, he spent fourteen years at Sun Microsystems, a now-defunct computer company that Bryan's nine-year-old daughter apparently thought was a brewery.
New Ways to Find Latency in Linux Using Tracing
Steven Rostedt currently works for VMware in their Open Source Technology Center. He's the maintainer of the PREEMPT_RT (aka Real Time patch) stable releases. He is also one of the original developers for the Real Time patch. Steven is the main developer and maintainer for ftrace, the official tracer of the Linux kernel, as well as the user space tools trace-cmd and kernelshark. He also develops ktest.pl (in the kernel) and was the creator of "make localmodconfig".
I/O Rings and You – Optimizing I/O on Windows
Yarden is a Software Engineer at Crowdstrike, working on EDR features, and a consultant for Winsider Seminars & Solutions Inc., co-teaching security trainings. Previously, she worked at SentinelOne as a security researcher and QA engineer. Outside of her primary work duties, Yarden writes articles and tools and gives talks and workshops about various topics such as CET internals, extension host hooking and kernel exploit mitigations. Outside of infosec, Yarden is a circus artist, teaching and performing aerial arts.
Data Structures for High Resolution, Real-time Telemetry at Scale
Filipe Oliveira is a Performance Engineer at Redis, working on both the enterprise and the open-source products as part of the CTO research group. Before Redis, he worked both on a large scale retail company and as Grid and Distributed Computing Researcher recurring to advanced computing infrastructures listed in the TOP500 supercomputers list.
Keeping Latency Low and Throughput High with Application-level Priority Management
Avi Kivity, CTO of ScyllaDB, is known mostly for starting the Kernel-based Virtual Machine (KVM) project, the hypervisor underlying many production clouds. He has worked for Qumranet and Red Hat as KVM maintainer until December 2012. Avi is now CTO of ScyllaDB, a company that seeks to bring the same kind of innovation to the public cloud space.
Get Lower Latency and Higher Throughput for Java Applications
Simon Ritter is the Deputy CTO of Azul, provider of the Java platform for the modern cloud enterprise. Simon joined Sun Microsystems in 1996 and started working with Java technology from JDK 1.0; he has spent time working in both Java development and consultancy. Having moved to Oracle as part of the Sun acquisition, he managed the Java Evangelism team for the core Java platform, Java for client applications and embedded Java. Now at Azul, he continues to help people understand Java as well as Azul’s JVM technologies and products. Simon has twice been awarded Java Rockstar status at JavaOne and is a Java Champion. He currently represents Azul on the JCP Executive Committee and on the Java SE Expert Group.
Rust Is Safe. But Is It Fast?
Glauber is a veteran low-level engineer with a strong focus on performance and resource management. He has worked with a variety of subsystems in the Linux Kernel, most notably the KVM Hypervisor and the cgroups resource management infrastructure that created the foundations of the containers revolution. Glauber had spent many years with ScyllaDB working with both business and technical issues, specializing in storage I/O and automated resource controlling. Currently he is a Staff Engineer at Datadog, where he authored the Glommio asynchronous framework for Rust.
Performance Analysis and Troubleshooting Methodologies for Databases
Peter Zaitsev is the co-founder and CEO of Percona, a provider of open-source enterprise-class MySQL, MongoDB, MariaDB and PostgreSQL software, solutions, and services. Co-author of High Performance MySQL: Optimization, Backups, and Replication, and the ebook Practical MySQL Performance Optimization, Zaitsev blogs regularly on the Percona Database Performance Blog and speaks frequently at conferences worldwide. Peter lives in North Carolina with his wife and two children and enjoys travel and spending time outdoors.
Extreme HTTP Performance Tuning: 1.2M API req/s on a 4 vCPU EC2 Instance
Marc is the resident Jack of All Trades at Talawah Solutions, a cloud computing consultancy that has served clients across the globe. A generalist at heart, Marc has worn a number of hats over the years, ranging from Analyst to Architect. Most recently, he has been indulging his passion for performance engineering in his attempts to explore the limits of HTTP performance in the cloud.
High-Performance Networking Using eBPF, XDP, and io_uring
Bryan is a Senior Software Engineer at Couchbase Inc. He currently works primarily in distributed systems on Couchbase Server but has always had a passion for low level, systems programming with an emphasis on Linux, async I/O, and (fast) networking in general. Proud member of multiple Linux Kernel Mailing lists (ebpf, io_uring, xdp). Rust language enthusiast, and overall lover of types and compilers. Currently an active contributor to the Glommio thread-per-core asynchronous runtime for the Rust programming language and supporter of all things open source. He has worked in many industries (eCommerce, Security, Databases) but always gravitated towards systems programming, no matter where that was.
OSNoise Tracer: Who Is Stealing My CPU Time?
Daniel is a Principal Software Engineer at Red Hat, working in the real-time kernel team. He has a Ph.D. in embedded real-time systems and in automation engineering. He is a post-doctoral researcher at the Retis Lab - Scuola Superiore Sant'Anna.
RISC-V on Edge: Porting EVE and Alpine Linux to RISC-V
Roman is a well known and acknowledged expert and visionary in Open Source strategy and execution. When co-founding ZEDEDA, he put open source technology at the core of a new industrial computing paradigm -- securing edge compute hardware as if it were part of your cloud-native ecosystem. In addition to making contributions to open source projects across a wide spectrum of technology, including active contributions to Alpine Linux and Project EVE, Roman is on the board of directors of LF Edge, and is VP Legal at the Apache Software Foundation.
Avoiding Data Hotspots at Scale
Kostja has started his database career in 2002 working on one of the first distributed genome databases helping scientists glue together partial genome sequences. He was a long time contributor of MySQL, Tarantool, and now, with ScyllaDB, where he's working on lightweight transactions and Raft replication protocol.
Crimson: Ceph for the Age of NVMe and Persistent Memory
Orit is an OpenShift Container Storage Architect at Red Hat, focusing on storage for containers, hybrid cloud, multi-cloud and Edge. An experienced software engineer who is passionate about open source and infrastructure, Orit has worked with distributed systems and storage for much of her career. She was a principal architect at Lightbits labs, a software-defined storage startup developing NVMe/TCP. Her journey at Red Hat started on Ceph object storage (Ceph Rados Gateway), a highly available distributed software-defined storage solution designed to scale and handle PetaBytes. She also co-maintained and developed live migration for KVM/QEMU. In the past at IBM Research Labs, she developed nested virtualization for KVM.
OSv Unikernel — Optimizing Guest OS to Run Stateless and Serverless Apps in the Cloud
Waldek Kozaczuk is a software developer and architect. During the day he is busy designing and building components of the video supply chain for CNN at WarnerMedia. During the night he is busy improving OSv, an open-source versatile modular unikernel designed to run a single unmodified Linux application securely on top of a hypervisor. He has been contributing to the project ince 2016, when he first got exposed to and fascinated by OSv. Three years ago he was nominated as one of its committers. In that time, he has contributed over 500 patches, which most notably involved enhancing OSv to run on AWS firecracker (http://blog.osv.io/blog/2019/04/19/making-OSv-run-on-firecraker/) and improving ARM port.
How to Measure Latency
Heinrich spent ~10 years of his live pursuing an academic career in Mathematics. After transitioning to IT, he worked as a Data Scientist for the Monitoring vendor Circonus and is now SRE @ Zalando.de. He has a passion understand the production behavior of IT systems, and tries to apply rigorous math and probabilistic modeling to this domain. Heinrich lives with his wife and 3 daughters in the rural town of Stemwede Germany.
Continuous Go Profiling
Felix Geisendörfer is a Staff Engineer at Datadog, where he works on Continuous Profiling for Go. Before that he was working on manufacturing systems for Apple, herding big PostgreSQL clusters. In his spare time he's usually working on [open source](https://github.com/felixge), playing beach volleyball, or giving [questionable advice](https://gist.github.com/felixge/024827e7f09048d7bee2ad7c397e6ace) on the internet.
Continuous Performance Regression Testing with JfrUnit
Gunnar Morling is a software engineer and open-source enthusiast by heart. He is leading the Debezium project, a platform for change data capture (CDC). He is a Java Champion, the spec lead for Bean Validation 2.0 (JSR 380) and has founded multiple open source projects such as Deptective and MapStruct. Prior to joining Red Hat, Gunnar worked on a wide range of Java EE projects in the logistics and retail industries. He's based in Hamburg, Germany.
Unikraft: Fast, Specialized Unikernels the Easy Way
Felipe Huici is chief researcher in the systems group at NEC Laboratories Europe in Heidelberg, Germany. His main research and work interests lie in the areas of high-performance software systems, and in particular specialization, virtualization, and the application of machine learning techniques to tackle open problems in the systems area. Previously, he received an undergraduate degree with honours from the University of Virginia, a Masters in Data Communications, Networks and Distributed Systems from University College London (top of the class), and a Ph.D. also from UCL. He has published on several top-tier conferences and journals such as SOSP, SIGCOMM, NSDI, CoNEXT, and SIGCOMM CCR and regularly act as TPC member of conferences and journals such as IMC , INFOCOM, CoNEXT and SIGCOMM CCR.
Seastore: Next Generation Backing Store for Ceph
Sam began working on the Ceph project in 2011. Most of his time currently is spent working on crimson, the next generation ceph-osd.
RISC-V on Edge: Porting EVE and Alpine Linux to RISC-V
Kathy fills her desk with hardware, especially compute nodes that can be flashed with EVE-OS and microcontroller boards that can be programmed as IoT devices. At ZEDEDA, she drives open source and industry collaboration for win-win benefits. Prior to ZEDEDA, she held senior roles at Mozilla, Arduino, Qualcomm, a few startups, and more. She also volunteers to teach students in STEM. She received her bachelor's degree in Electrical Engineering from the University of Minnesota, and her master's in EE from Stanford University. When not at a computer, she can usually be found exploring the great outdoors of northern California.
Realtime Indexing for Fast Queries on Massive Semi-Structured Data
Dhruba Borthakur is CTO and co-founder of Rockset, responsible for the company's technical direction. He was an engineer on the database team at Facebook, where he was the founding engineer of the RocksDB data store. Earlier at Yahoo, he was one of the founding engineers of the Hadoop Distributed File System. He was also a contributor to the open source Apache HBase project. Dhruba previously held various roles at Veritas Software, founded an e-commerce startup, Oreceipt.com, and contributed to Andrew File System (AFS) at IBM-Transarc Labs.
DB Latency Using DRAM + PMem in App Direct & Memory Modes
Obsessed with Database performance and availability. Developer and product manager for Oracle TimesTen In-Memory Database.
Vanquishing Latency Outliers in the Lightbits LightOS Software Defined Storage System
Abel Gordon is the Chief System Architect of Lightbits Labs, where he architected a low latency clustered block storage solution based on NVMe/TCP protocol. Abel has more than 20 years of experience researching, designing and building systems, and has published more than 40 papers and patents. Prior to joining Lightbits labs, Abel led the Kernel and Networking team of Stratoscale's hyper-converged private cloud solution. Abel also worked at IBM Research -- Haifa, where he led multiple activities in the area of machine and I/O virtualization. He contributed new features to the KVM hypervisor and co-authored award-winning papers. Abel also designed and implemented modern software systems to control unmanned aerial vehicles while he worked at Elbit Systems. Abel holds a B.Sc. (Cum Laude) in Information Systems Engineering from the Technion - Israel Institute of Technology.
Where Did All These Cycles Go?
Thomas spent 20+ years in computer security (under the pseudonym ""Halvar Flake""); he built a company around low-level code analysis tools for analyzing security patches and malware which was acquired by Google in 2011. He also worked in various realms of security research, among other things into bad DRAM (RowHammer). At some point, he realized that performance work is very similar to security work: In security, you analyze legacy systems, find issues, and then everybody is angry -- but in performance, you can analyze legacy systems, find problems, and then everybody is happy because you made it faster. As a consequence, he co-founded optimyze.cloud in 2018, where he works on building large-scale performance observability and optimization tooling.
Using SLOs for Continuous Performance Optimizations of Your k8s Workloads
Andreas Grabner is a contributor to the CNCF open source project Keptn. He has been working in performance engineering for the past 20 years and is sharing his experiences via blogs, videos or his PurePerformance podcast. You can follow him via @grabnerandi
Let’s Fix Logging Once and for All
Performance Engineer working at Red Hat, Inc for the past 9+ years on logging and general performance issues. Prior to that spent 20 years at HP/Compaq/Digital on TruClusters, gWLM, POSIX Threads, and OpenVMS print symbiont.
What We Need to Unlearn about Persistent Storage
Pavel Emelyanov, principal engineer, ex Linux kernel hacker now speeding up row cache, tweaking IO scheduler and helping to pay back a tech debt for components innter-dependencies.
How to Meet Your P99 Goal While Overcommitting Another Workload
Shlomi Livne has 15 years of experience in software development of large scale systems. Previously he has led the research and development team at Convergin, which was acquired by Oracle. Shlomi holds a BA and MSc in Computer Science from the Technion-Machon Technologi Le’ Israel.
Is It Faster to Go with Redpanda Transactions than Without Them?!
Denis Rystsov is a staff engineer at Vectorized working on Redpanda, a Kafka-compatible distributed event streaming platform. His area of expertise lies in distributed transactions, consensus algorithms, and testing consistency models. He likes to keep an eye on academic computer science research and use it as leverage in the industry. Denis volunteers as a member of the Hydra conference program committee and as a member of the editorial board of the Journal of Systems Research (JSys). He received his MSc in Applied Mathematics and Computer Science from SPbSU. Before joining Vectorized he worked at Yandex, Amazon, and Microsoft. Apart from work he likes surfing and exploring the waters of the PNW.