Bio

I hate writing bios. From time to time, conference organizers, journalists, and publishers ask me for a bio of a certain length ("50-75 words, "around 100 words", etc.), and so I'll cache my most recent ones here. This page also has my upcoming talks.

Very Short

Industry expert in computing performance and eBPF. Solves hard problems. Makes things faster.

Short (<75 words)

Brendan Gregg is an internationally renowned expert in computing performance. He is a senior performance architect at Netflix, where he does performance design, evaluation, analysis, and tuning. He authored Systems Performance and BPF Performance Tools (Addison-Wesley), and received the USENIX LISA Outstanding Achievement award. He has delivered industry-leading performance for a variety of products, and has created widely used performance tools, methodologies, and visualizations, including flame graphs. His work is the basis for multiple startups.

Medium (around 100 words)

Brendan Gregg is an internationally renowned expert in computing performance. He is a senior performance architect at Netflix, where he does performance design, evaluation, analysis, and tuning. He authored Systems Performance and BPF Performance Tools in the Addison-Wesley professional computing series, and received the USENIX LISA Outstanding Achievement award. Previously among the top performance experts at Sun Microsystems, he has delivered industry-leading performance for a variety of products. He has also created widely used performance tools, methodologies, and visualizations, including flame graphs, and pioneered eBPF as an observability technology. His work has saved the industry over US$1B, and has been the basis for multiple startups.

Long

Brendan Gregg is an internationally renowned expert in computing performance. He is a senior performance architect at Netflix, where he does large-scale computer performance design, evaluation, analysis, and tuning. Previously among the top performance experts at Sun Microsystems, he has delivered industry-leading performance for a variety of commercial products and enterprise environments. He has also created widely used tools, methodologies, and visualizations for performance analysis, including flame graphs. His tools are included in multiple operating systems and products, and are in use by companies worldwide. He pioneered eBPF as an observability technology, and leads the industry in its usage. His work has saved the industry over US$1B through performance improvements, and has been the basis for multiple startups.

He has authored hundreds of articles about systems performance and multiple technical books, including Systems Performance 2nd Edition (2021) and BPF Performance Tools (2019), both in the Addison-Wesley professional computing series. His books are recommended or required reading at major technology companies. He has also been an invited keynote speaker at technical conferences worldwide, and received the USENIX LISA Outstanding Achievement award.

He provides performance expertise for multiple company functions, including directing other teams, evaluating new technologies, and making individual contributions. He has extensive performance tuning experience for enterprise and small- to large-scale cloud computing environments, and root cause analysis of complex problems across the software stack and down to metal. He has worked with sales organizations to win and retain customers during proof of concepts and product support, including direct work with Fortune 500 companies, securing millions in company revenue. He represented Sun Microsystems and Netflix worldwide as a performance expert.

His creations include flame graphs and latency heat maps, the USE methodology, and advanced eBPF tracing tools to provide unique insights into system behavior. His prior work includes leading performance efforts for the world's first container-based cloud, and for the first ZFS storage appliance, in both cases delivering industry-leading performance. As a kernel engineer, he developed the ZFS L2ARC: A pioneering file system performance technology. He has also developed and delivered professional training courses on computer performance.

Way too long make it stop

Brendan Gregg is an internationally renowned expert in computing performance. He is currently a senior performance architect at Netflix, where he does large-scale computer performance design, evaluation, analysis, and tuning. Previously among the top performance experts at Sun Microsystems, he has delivered industry-leading performance for a variety of commercial products and enterprise environments. Throughout his career he has also created widely used tools, methodologies, and visualizations for performance analysis, including flame graphs. His tools are included in multiple operating systems and products, and in use by companies worldwide. He has also pioneered eBPF as an observability technology, and leads the industry in its usage. His work has saved the industry over US$1B through performance improvements, and has been the basis for multiple startups.

He is the author of the books Systems Performance: Enterprise and the Cloud, 2nd Edition (Addison-Wesley, 2020), as well as 1st Edition (Prentice Hall, 2013), BPF Performance Tools: Linux System and Application Observability (Addison-Wesley, 2019), and is the primary author of DTrace: Dynamic Tracing in Oracle Solaris, Mac OS X, and FreeBSD (Prentice Hall, 2011), and co-author of Solaris Performance and Tools (Prentice Hall, 2006). He has also written hundreds of articles about systems performance. His books have been recommended or required reading at major technology companies. He is an in-demand conference talk and invited keynote speaker, and has spoken at over 60 technical conferences worldwide. He won the JavaOne rockstar speaker award in 2016, and the DockerCon top speaker award in 2017.

At Netflix, his work includes evaluating new technologies and opportunities, providing performance expertise to guide major changes and projects, and creating and using new advanced performance tools. This can involve any application, at any level of the stack, and for cloud instances and physical servers. He has developed and used analysis tools for the JDK hotspot and Node.js v8 runtimes, the Linux and FreeBSD kernels, the Xen hypervisor, Linux containers, and Docker. He has expertise with and has made contributions to multiple technologies, including Linux perf, Ftrace, eBPF, PMCs, ZFS, and DTrace. He has developed, debugged, and tuned software in rings 0, 1, and 3. He has also assisted production triage for performance and availability issues, and has been the primary on-call engineer on the Netflix CORE SRE (site reliability engineering) team rotation to lead the worldwide response for Netflix outages.

His previous roles have included kernel engineering, performance engineering, enterprise security, system administration, and programming. While working at Sun Microsystems, he led performance efforts for the world's first ZFS-based storage appliance, helping make it the fastest product of its class. As a kernel engineer he also developed the ZFS L2ARC – a pioneering file system performance technology which uses flash memory as a second-level cache, after main memory. (The original idea of a ZFS L2ARC was by Adam Leventhal for his hybrid storage pool vision.) After Sun, he was the lead performance engineer at a cloud computing startup, successfully beating AWS performance to win customers during proof of concepts, and to deliver the company's motto of a "high-performance cloud."

Throughout his career he has become a recognized name and world expert in different technologies to stay current with the ever-changing industry. Originally becoming known as one of the top experts in Solaris performance, he then became known as the expert for DTrace tools, ZFS performance, cloud performance, Linux tracing, and now eBPF observability.

For contributions to the field of system administration, particularly his work in systems performance analysis methodologies, he is the recipient of the USENIX 2013 LISA Award for Outstanding Achievement in System Administration. He later contributed to USENIX as a member of the LISA steering committee, and was program co-chair with Rikki Endsley for the USENIX LISA 2018 conference, in its 32nd year. LISA was one of the longest running technology conferences, and focused on rigorous engineering of large scale computer systems (LISA originally stood for Large Installation System Administration).

He has invented and developed numerous advanced performance analysis tools, methodologies, and visualizations. His eBPF-based performance tools are installed by default on servers at Netflix and Facebook. His DTrace-based performance tools are shipped by default in the Mac OS X, FreeBSD, and Oracle Solaris 11 operating systems. He has created the DTraceToolkit, the Ftrace-based perf-tools collection, and has also contributed code, documentation, testing, leadership, and performance tools to the eBPF-based bcc and bpftrace open source projects.

His recent work has included new performance methodologies and visualizations for Linux kernel analysis, and performance tools for both Linux and FreeBSD. His methodologies include the USE method, off-CPU analysis, and the TSA method, and his visualizations include latency, utilization, and sub-second offset heat maps, flame graphs, and flame scope.

Early in his career he worked as an independent performance consultant, hired by various companies to debug a wide variety of performance issues, and to provide custom training for staff.

He also previously worked as a professional technical trainer (instructor), teaching classes worldwide that were typically five days in length. For Sun Microsystems Educational Services he taught over a dozen different classes (an extraordinarily large portfolio), which if ran back-to-back would take three months of all-day training to complete. These classes included performance tuning, system administration, network administration, administering security, network intrusion detection, fault analysis, and shell programming. These were all subjects for which he had prior production experience. For Sun Microsystems he also became the lead instructor for a number of classes in APAC, including performance tuning. He has also authored and delivered his own technical workshops, including a 10-day advanced performance tuning class taught internally at Sun Microsystems in the US. He has taught over 2,000 people in the industry.

Born in Australia, he first worked in the Asia Pacific region, then spent many years working in Silicon Valley, California. He returned to Australia in 2020, and is now based in Sydney. (In case my bio inspires other Australians or non-US citizens to consider working in the US, then I'd strongly recommend that you get a green card beforehand, else consider remote work instead of a more restrictive visa.)

For selected examples of his work, see the overview page.

Interviews

Talks

Upcoming talks:

Conferences I will or have spoken at. Mouse-over for names (tooltip), click for websites:

Prior conference talks include (see my homepage for slides and video):

  1. "Performance Wins with BPF: Getting Started", Brendan Gregg, eBPF Summit, online 2021
  2. "Performance Wins with BPF: Getting Started", Brendan Gregg, Facebook @Scale, online 2021
  3. "Computer Performance: On the Horizon" (plenary), Brendan Gregg, USENIX LISA, online 2021
  4. "BPF Internals", Brendan Gregg, USENIX LISA, online 2021
  5. "Performance Wins with BPF: Getting Started" (keynote), Brendan Gregg, eBPF Summit, online 2020
  6. "Linux Systems Performance" (keynote), Brendan Gregg, YOW! Auckland, Perth, Singapore, Hong Kong, online 2020
  7. "Brendan/Netflix Updates", Brendan Gregg, LSFMMBPF, online 2020
  8. "BPF Performance Analysis at Netflix", Brendan Gregg, AWS re:Invent, Las Vegas 2019
  9. "Linux Systems Performance", Brendan Gregg, USENIX LISA, Portland 2019
  10. "Extended BPF: A New Type of Software" (keynote), Brendan Gregg, UbuntuMasters, Santa Clara 2019
  11. "BPF Observability" (keynote), Brendan Gregg, LSFMM, San Juan, Puerto Rico 2019
  12. "eBPF Performance Tools 2019", Brendan Gregg, SCaLE17x, Los Angeles 2019
  13. "Cloud Performance Root Cause Analysis at Netflix" (keynote), Brendan Gregg, YOW! Conference, Sydney & Brisbane & Melbourne, Australia 2018
  14. "Working at Netflix" (keynote), Brendan Gregg, YOW! CTO Summit, Sydney & Melbourne, Australia 2018
  15. "Linux Performance 2018", Brendan Gregg, All Things Open, Raleigh 2018
  16. "BPF Observability", Brendan Gregg, NetConf, Boston 2018
  17. "Linux Performance 2018" (keynote), Brendan Gregg, PerconaLive, Santa Clara 2018
  18. "CPU Utilization is WRONG" (lightning talk), Brendan Gregg, SCALE16x, Los Angeles 2018
  19. "How Netflix Tunes EC2 Instances for Performance", Brendan Gregg, AWS re:Invent, Las Vegas 2017
  20. "Linux Container Performance Analysis", Brendan Gregg, USENIX LISA, San Francisco 2017
  21. "Using Linux perf at Netflix", Brendan Gregg, Kernel Recipes, Paris 2017
  22. "Performance Analysis Superpowers with Linux BPF", Brendan Gregg, Kernel Recipes, Paris 2017
  23. "System Performance Analysis Methodologies" (keynote), Brendan Gregg, EuroBSDcon, Paris 2017
  24. "Performance Analysis Superpowers with Linux BPF", Brendan Gregg, Open Source Summit, Los Angeles 2017
  25. "Performance Superpowers with Enhanced BPF", Brendan Gregg, USENIX ATC, Santa Clara 2017
  26. "Visualizing Performance with Flame Graphs", Brendan Gregg, USENIX ATC, Santa Clara 2017
  27. "Performance Analysis Superpowers with Linux eBPF", Brendan Gregg, O'Reilly Velocity, San Jose 2017
  28. "Container Performance Analysis", Brendan Gregg, DockerCon, Austin 2017
  29. "Linux 4.x Tracing: Performance Analysis with bcc/BPF", Brendan Gregg, SCALE15x, Los Angeles 2017
  30. "BPF Tools", Brendan Gregg, IO Visor Summit, Mountain View 2017
  31. "Linux Monitoring at Scale with eBPF", Alex Maestretti, Brendan Gregg, BsidesSF, San Francisco 2017
  32. "BPF: Tracing and More", Brendan Gregg, linux.conf.au, Hobart, Australia 2017
  33. "Linux 4.x Tracing Tools: Using BPF Superpowers", Brendan Gregg, USENIX LISA, Boston 2016
  34. "Designing Tracing Tools", Brendan Gregg, Sysdig CCWFS, San Francisco 2016
  35. "Java Performance Analysis on Linux with Flame Graphs", Brendan Gregg, JavaOne, San Francisco 2016
  36. "System Methodology - Holistic Performance Analysis on Modern Systems" (keynote), Brendan Gregg, ACM Applicative, New York 2016
  37. "Linux Systems Performance", Brendan Gregg, PerconaLive, Santa Clara 2016
  38. "Performance Checklists for SREs" (keynote), Brendan Gregg, USENIX SREcon, Santa Clara 2016
  39. "Linux 4.x Performance: Using BPF Superpowers", Brendan Gregg, Performance@Scale, Menlo Park 2016
  40. "Broken Linux Performance Tools", Brendan Gregg, SCALE14x, Los Angeles 2016
  41. "Broken Performance Tools", Brendan Gregg, QCon, San Francisco 2015
  42. "Java Mixed-Mode Flame Graphs", Brendan Gregg, JavaOne, San Francisco 2015
  43. "Netflix Instance Performance Analysis Requirements", Brendan Gregg, Monitorama, Portland 2015
  44. "Linux Profiling at Netflix", Brendan Gregg, SCALE13x, Los Angeles 2015
  45. "Linux Performance Analysis: New Tools and Old Secrets", USENIX LISA, Seattle 2014
  46. "Performance Tuning EC2 Instances", Brendan Gregg, AWS re:Invent, Las Vegas 2014
  47. "FreeBSD Flame Graphs", Brendan Gregg, FreeBSD Developer and Vendor Summit, San Jose 2014
  48. "Performance Analysis", Brendan Gregg, MeetBSD California, San Jose 2014
  49. "Linux Performance Tools", Brendan Gregg, LinuxCon | CloudOpen Europe, Düsseldorf 2014
  50. "From DTrace to Linux", Brendan Gregg, Tracing Summit, Düsseldorf 2014
  51. "From Clouds to Roots", Brendan Gregg, SurgeCon, Baltimore 2014
  52. "Linux Performance Tools", Brendan Gregg, LinuxCon | CloudOpen North America, Chicago 2014
  53. "Analyzing OS X Systems Performance with the USE Method", Brendan Gregg, MacIT, San Francisco 2014
  54. "What Linux can learn from Solaris performance and vice-versa" (keynote), Brendan Gregg, SCaLE12x, Los Angeles 2014
  55. "Blazing Performance with Flame Graphs" (plenary), Brendan Gregg, USENIX LISA, Washington DC, 2013
  56. "Open Source Systems Performance", Brendan Gregg, O'Reilly Open Source Convention (OSCON), Portland 2013
  57. "Stop the Guessing: Performance Methodologies for Production Systems", Brendan Gregg, Velocity, Santa Clara 2013
  58. "Linux Performance Analysis and Tools", Brendan Gregg, SCaLE11x, Los Angeles 2013
  59. "Performance Analysis Methodology", Brendan Gregg, USENIX LISA, San Diego 2012
  60. "ZFS Performance Analysis and Tools", Brendan Gregg, zfsday, San Francisco 2012
  61. "DTracing the Cloud", Brendan Gregg, illumosday, San Francisco 2012
  62. "The USE Method", Brendan Gregg, Fórum Internacional Software Livre 13 (FISL13), Porto Alegre 2012
  63. "Visualizations", Brendan Gregg, dtrace.conf, San Francisco 2012
  64. "Performance Analysis: new tools and concepts from the cloud", Brendan Gregg, Southern California Linux Expo (SCaLE10x), Los Angeles 2012
  65. "Breaking Down Query Latency With DTrace", Brendan Gregg, Percona Live, New York 2011
  66. "Visualizations for Performance Analysis", Brendan Gregg, Large Installation System Admininistration (USENIX LISA), San Jose 2010
  67. "How To Build Better Applications With Oracle Solaris DTrace", Brendan Gregg, Jim Mauro, Oracle Open World (OOW), San Francisco 2010
  68. "DTrace Crash Course", Brendan Gregg, Kernel Conference Australia (KCA), Brisbane 2009
  69. "Kernel Engineering", Brendan Gregg, Kernel Conference Australia (KCA), Brisbane 2009

Prior tutorials/workshops:

  1. "BPF Performance Tools", Brendan Gregg, USENIX LISA, Portland 2019
  2. "Linux Performance Analysis Using Perf and BPF", Brendan Gregg, Sasha Goldshtein, USENIX LISA, Boston 2016
  3. "Linux Performance Tools", Brendan Gregg, O'Reilly Velocity, Santa Clara 2015
  4. "Metrics Workshop", Brendan Gregg, Narayan Desai, Kent Skaar, Theo Schlossnagle, Caskey Dickson, USENIX LISA, Washington DC 2013
  5. Not listed here are instructor led professional training classes I've developed and delivered, which ranged from 3 to 10 days in length.

Last updated: 19-Jul-2021