Dynatrace

MAY 26, 2021

From May 17 to May 18, 2021, the Open-Source Engineering team at Dynatrace attended the virtual observability conference, o11yfest. The two-day conference offered two tracks: One dedicated to the business value of OpenTelemetry observability, where we got an insight into use cases from Skyscanner, HBE Digital, and Logz.io, to name a few.

AWS 221

AWS Strategy Storage Metrics 221

Brendan Gregg

FEBRUARY 28, 2023

At USENIX SREcon22 APAC I gave the opening keynote on the future of computer performance, rounding up the latest developments and making predictions of where I see things heading. This talk originated from my updates to Systems Performance 2nd Edition , and this was the first time I've given this talk in person!

Performance 52

Performance Latency Cache Virtualization 52

Brendan Gregg

JULY 4, 2021

It's an exciting time for developments in computer performance, not just for the BPF technology (which I often [write about]) but also for processors with 3D stacking and cloud vendor CPUs (e.g., This was a chance to talk about other things I've been working on, such as the present and future of hardware performance.

Performance 52

Performance Latency Hardware Storage 52

All Things Distributed

SEPTEMBER 20, 2013

Not everybody agreed that the "N-ary Storage Model" (NSM) was the best approach for all workloads but it stayed dominant until hardware constraints, especially on caches, forced the community to revisit some of the alternatives. Many of the modern high-performance data warehouses such as Amazon Redshift are based on column stores.

Storage 87

Storage Hardware Cache C++ 87

Dynatrace

MAY 26, 2021

From May 17 to May 18, 2021, the Open-Source Engineering team at Dynatrace attended the virtual observability conference, o11yfest. The two-day conference offered two tracks: One dedicated to the business value of OpenTelemetry observability, where we got an insight into use cases from Skyscanner, HBE Digital, and Logz.io, to name a few.

AWS 147

AWS Strategy Storage Metrics 147

Brendan Gregg

APRIL 14, 2022

On the Netflix Java/Linux/EC2 stack there were no working mixed-mode flame graphs, no production safe dynamic tracer, and no PMCs: All tools I used extensively for advanced performance analysis. Apart from developing tools, much of my time has been spent helping teams with performance issues and evaluations. How would I do my job?

Java 145

Java Cloud Cache AWS 145

All Things Distributed

DECEMBER 5, 2019

During AWS re:Invent 2019, we announced a number of High Performance Computing (HPC) innovations including the Amazon EC2 M6g, C6g, and R6g instances powered by next-generation Arm-based AWS Graviton2 Processors. We also recently announced that new AMD-powered, compute-optimized EC2 instances are in the works.

AWS 80

AWS Storage Innovation Network 80

All Things Distributed

OCTOBER 2, 2017

We were pushing the limits of what was a leading commercial database at the time and were unable to sustain the availability, scalability and performance needs that our growing Amazon business demanded. Performant – The service would need to be able to maintain consistent performance in the face of diverse customer workloads.

Internet 128

Internet Internet AWS Performance 128

Percona

NOVEMBER 1, 2023

Learn more: Discover six of the most common causes of poor database performance in our free eBook. Offsite Storage It’s highly recommended to copy all the backup methods to another place, like the cloud or an external file server, so in case of host failure or data center failure, you have another copy. ” Great advice.

Best Practices 78

Best Practices Storage Strategy Database 78

All Things Distributed

MAY 19, 2017

In many, high-throughput, OLTP style applications the database plays a crucial role to achieve scale, reliability, high-performance and cost efficiency. We believe the central constraint in high throughput data processing has moved from compute and storage to the network.

Database 117

Database Design Cloud Storage 117

High Scalability

APRIL 3, 2019

We found the answers at the Postgres Conference in March where we surveyed PostgreSQL users, contributors, and SQL and NoSQL database administrators alike. For the 2nd year in a row, PostgreSQL has kept the title of #1 fastest growing database in the world according to the DBMS of the Year report by the experts at DB-Engines.

Database 177

Database Cloud Open Source Systems 177

The Morning Paper

NOVEMBER 8, 2020

USENIX is a nonprofit organisation committed to making content and research freely available – both conference proceedings and the recorded presentations of their events. Without in-person conferences this year, income is down and events are under threat. What Facebook needed, literally, was a plan to throw one away.

Virtualization 126

Virtualization Latency Storage Systems 126

Dynatrace

MARCH 23, 2020

A decade ago, while working for a large hosting provider, I led a team that was thrown into turmoil over the purchasing of server and storage hardware in preparation for a multi-million dollar super-bowl ad campaign. I gave an early glimpse of the Google integration last month at our annual Perform user conference in Las Vegas.

Google 196

Google Engineering Metrics Hardware 196

Smashing Magazine

JANUARY 15, 2019

How Improving Website Performance Can Help Save The Planet. How Improving Website Performance Can Help Save The Planet. For the last three years, the SustainableUX conference has seen experts in web sustainability sharing their knowledge across an array of web-based disciplines. Jack Lenox. 2019-01-15T13:30:32+01:00.

Website Performance 73

Website Performance Website Energy Performance 73

Brendan Gregg

APRIL 14, 2022

On the Netflix Java/Linux/EC2 stack there were no working mixed-mode flame graphs, no production safe dynamic tracer, and no PMCs: All tools I used extensively for advanced performance analysis. Apart from developing tools, much of my time has been spent helping teams with performance issues and evaluations. How would I do my job?

Java 52

Java Cache Cloud AWS 52

The Morning Paper

DECEMBER 8, 2019

At the start of November I was privileged to attend HPTS (the High Performance Transaction Systems) conference in Asilomar. On the last morning of the conference Daniel Bittman presented some of the work being done in the context of the Twizzler OS project to explore new programming models for NVM. Early performance results.

Hardware 56

Hardware Programming Media Storage 56

The Netflix TechBlog

NOVEMBER 3, 2022

When a new leader is elected it loads all data from external storage. In USENIX annual technical conference (Vol. Overview The figure below depicts a simplified high-level architecture of a single Titus cluster (a.k.a cell): Titus Job Coordinator is a leader elected process managing the active state of the system. 1996, January).

Cache 224

Cache Latency Traffic Systems 224

The Morning Paper

NOVEMBER 30, 2020

Our approach is premised on a single observation: applications view storage systems as black-boxes that transition through a series of states, a subset of which are observed by applications. A storage system supports read and write operations over a set of keys $K$. Revealing performance enhancement opportunities.

Database 61

Database Storage Systems Strategy 61

Brendan Gregg

APRIL 30, 2018

USENIX’s LISA conference is the premier event for topics in production system engineering. Another exciting change for 2018: LISA is now a 3-day event, rather than 6 days; check out the Announcement of Changes to Future LISA Conferences for more details. Join us for 3 days in Nashville at LISA'18. Photo by Perry Kivolowitz.

DevOps 43

DevOps Network Best Practices Programming 43

All Things Distributed

DECEMBER 18, 2012

After the AWS re: Invent conference I spent two weeks in Europe for the last customer visits of the year. Aug 24 - The 5 Minute Rule for Trading Memory for Disk Accesses and The 10 Byte Rule for Trading Memory for CPU Time , Jim Gray and Gianfranco Putzolu, Proceedings of the ACM SIGMOD Conference, pp. Comments (). Howard, Michael L.

Operating System 144

Operating System Cache AWS Scalability 144

Brendan Gregg

APRIL 29, 2018

USENIX’s LISA conference is the premier event for topics in production system engineering. Another exciting change for 2018: LISA is now a 3-day event, rather than 6 days; check out the Announcement of Changes to Future LISA Conferences for more details. Join us for 3 days in Nashville at LISA'18. Photo by Perry Kivolowitz.

DevOps 40

DevOps Network Best Practices Programming 40

The Symphonia

JUNE 19, 2018

Some in-depth and unusual client work (want to encrypt a petabyte of S3 storage? We’ll get to all of those later on, but first I’m going to start the news this time with a roundup of an interesting day last week… News from the Serverless World Keynote Stage at Velocity 2018 Last week I was at O’Reilly’s Velocity conference in San Jose.

Serverless 40

Serverless AWS DevOps Open Source 40

Adrian Cockcroft

JUNE 21, 2022

I became the Sun UK local specialist in performance and hardware, and as Sun transitioned from a desktop workstation company to sell high end multiprocessor servers I was helping customers find and fix scalability problems. Rich became co-author of the second edition of the Sun Performance Tuning book, to describe how it worked.

Tuning 52

Tuning Benchmarking Engineering C++ 52

The Morning Paper

SEPTEMBER 19, 2019

There were so many interesting papers at the conference this year though that I haven’t been able to cover nearly as many as I would like. Autoscaling tiered cloud storage in Anna. The authors claim a two orders-of-magnitude performance improvement. speedup over the best performing existing method. Research papers. (In

Blockchain 46

Blockchain Hardware Google Analytics 46

All Things Distributed

DECEMBER 5, 2019

During AWS re:Invent 2019, we announced a number of High Performance Computing (HPC) innovations including the Amazon EC2 M6g, C6g, and R6g instances powered by next-generation Arm-based AWS Graviton2 Processors. We also recently announced that new AMD-powered, compute-optimized EC2 instances are in the works.

AWS 17

AWS Storage Innovation Network 17

The Morning Paper

MAY 27, 2019

Last week we saw the benefits of rethinking memory and pointer models at the hardware level when it came to object storage and compression ( Zippads ). That ‘reasonable’ performance cost is a 6.8% total memory, and around 50% performance). ASPLOS’19. We answer this question affirmatively. Conceptual model. slower to 9.8%

C++ 61

C++ Hardware Virtualization Benchmarking 61

All Things Distributed

OCTOBER 2, 2007

In two weeks we’ll present a paper on the Dynamo technology at SOSP, the prestigious biannual Operating Systems conference. Dynamo is internal technology developed at Amazon to address the need for an incrementally scalable, highly-available key-value storage system.

Operating System 80

Operating System Storage Scalability Technology 80

ScaleOut Software

JUNE 29, 2018

Attending technical conferences creates the opportunity to step away from focusing on day-to-day concerns and reflect more deeply about the key principles that guide our work. However, this model significantly simplifies application development and enables the platform to maximize stream-processing performance. By Dr. William L.

Software Architecture 40

Software Architecture Architecture Software Software 40

ScaleOut Software

JUNE 29, 2018

Attending technical conferences creates the opportunity to step away from focusing on day-to-day concerns and reflect more deeply about the key principles that guide our work. However, this model significantly simplifies application development and enables the platform to maximize stream-processing performance. By Dr. William L.

Software Architecture 40

Software Architecture Architecture Software Software 40

The Morning Paper

SEPTEMBER 5, 2019

The ability to compute quantiles over aggregated metrics has been recognized to be an essential feature of any monitoring system… Given how expensive calculating exact quantiles can be for both storage and network bandwidth, most monitoring system will compress the data into sketches and compute approximate quantiles.

Latency 52

Latency Metrics Open Source Java 52

ACM Sigarch

NOVEMBER 4, 2019

Conference. The best paper runner-up was “ Dynamic Multi-Resolution Data Storage ”. . The lightning talk sessions on Monday and Tuesday helped the audience decide which talks to attend, and the poster sessions also provided more interactions between the authors and conference attendees. . Workshops and Tutorials.

Hardware 18

Hardware Architecture Programming Innovation 18

ACM Sigarch

NOVEMBER 1, 2019

This blog post gives a glimpse of the computer systems research papers presented at the USENIX Annual Technical Conference (ATC) 2019, with an emphasis on systems that use new hardware architectures. USENIX ATC is a top-tier venue with a broad range of systems research papers from both industry and academia. Heterogeneous ISA.

Architecture 12

Architecture Hardware Cache Storage 12

Brendan Gregg

SEPTEMBER 5, 2017

I gave a talk about container performance recently at DockerCon, and included a quick summary of how they work: YouTube , SlideShare : If you search of slideshare and youtube, you'll find many other good talks on containers as well. People rarely create them manually.

Virtualization 76

Virtualization AWS Engineering Hardware 76

Adrian Cockcroft

NOVEMBER 11, 2019

It would take 5–10 minutes after the problem occurred to appear as a problem, then people have to notice and respond to emails or pager text messages, dial into a conference call, and log in to monitoring dashboards before any human response can start. The third technique is lineage driven fault detection [Paper: Peter Alvaro].

Latency 52

Latency Engineering Systems Hardware 52

Adrian Cockcroft

NOVEMBER 11, 2019

It would take 5–10 minutes after the problem occurred to appear as a problem, then people have to notice and respond to emails or pager text messages, dial into a conference call, and log in to monitoring dashboards before any human response can start. The third technique is lineage driven fault detection [Paper: Peter Alvaro].

Latency 53

Latency Engineering Systems Hardware 53