Revealing Hidden Latency Patterns

Latency Heat Map Response time – or latency – is crucial to understand in detail, but many of the common presentations of this data hide important details and patterns. Latency heat maps are an effective way to reveal these. I often use tools that provide heat maps directly, but sometimes I have separate trace output [...]

Read more...
Posted on May 19, 2013 at 2:56 pm by Brendan Gregg · Permalink · Comments Closed
In: heatmaps, latency, performance, visualizations

Delphix and Flash

I started working with flash in 2006 — fortunate timing as flash was just starting to take hold in the enterprise. I started asking customers I’d visit about flash. I’ll always remember the response from an early adopter when I asked about how he planned on using the new, expensive storage, “We just bought it, [...]

Read more...
Posted on May 5, 2013 at 9:28 pm by ahl · Permalink · Comments Closed
In: Delphix, flash

Agile Data Technology

Applications are the nexus of the modern enterprise. They simplify operations, speed execution, and drive competitive advantage. Accelerating the application lifecycle means accelerating the business. Increasingly, organizations turn to public and private clouds, SaaS offerings, and outsourcing to hasten development and reduce risk, only to find themselves held hostage by their data. Applications are nothing [...]

Read more...
Posted on April 21, 2013 at 1:00 pm by eschrock · Permalink · Comments Closed
In: Delphix

Enterprise Software Hackathons

At Delphix, we just concluded one of our recurring Engineering Kickoff events where we get everyone together for a few days of collaboration, discussion, idea sharing, and fun. In this case it included, for the first time, an all-day hackathon event. To be honest, it was a bit of an experiment and one where we [...]

Read more...
Posted on February 28, 2013 at 2:43 pm by eschrock · Permalink · Comments Closed
In: Delphix

On Systems Software

A prospective new college hire recently related an odd comment from his professor: systems programming is dead. I was nonplussed; what could the professor have meant? Systems is clearly very much alive. Interesting and important projects march under the banner of systems. But as I tried to construct a less emotional rebuttal, I realized I [...]

Read more...
Posted on February 24, 2013 at 9:46 pm by ahl · Permalink · Comments Closed
In: software, systems

The Holistic Engineer

The idea of the holistic engineer embodies the point of view that an engineer needs to consider the whole system, the whole body of work that makes a product successful. It bears no relation to holistic health — and it’s not some even newer age quackery. There are many specialist roles in the software industry — [...]

Read more...
Posted on February 6, 2013 at 1:02 am by ahl · Permalink · Comments Closed
In: software

Virtualization Performance: Zones, KVM, Xen

At Joyent we run a high-performance public cloud based on two different virtualization technologies: Zones and KVM. We have historically run Xen as well, but have phased it out for KVM on SmartOS. My job is to make things go fast, which often means using DTrace to analyze the kernel, applications, and those virtualization technologies. [...]

Read more...
Posted on January 11, 2013 at 4:58 pm by Brendan Gregg · Permalink · Comments Closed
In: Cloud, DTrace, KVM, performance, xen, zones

zfsday: ZFS Performance Analysis and Tools

At zfsday 2012, I gave a talk on ZFS performance analysis and tools, discussing the role of old and new observability tools for investigating ZFS, including many based on DTrace. This was a fun talk – probably my best so far – spanning performance analysis from the application level down through the kernel and to [...]

Read more...
Posted on December 29, 2012 at 6:04 pm by Brendan Gregg · Permalink · Comments Closed
In: performance, slides, talk, video, ZFS

The USE Method: SmartOS Performance Checklist

The USE Method provides a strategy for performing a complete check of system health, identifying common bottlenecks and errors. For each system resource, metrics for utilization, saturation and errors are identified and checked. Any issues discovered are then investigated using further strategies. In this post, I’ll provide an example of a USE-based metric list for [...]

Read more...
Posted on December 19, 2012 at 10:23 am by Brendan Gregg · Permalink · Comments Closed
In: illumos, omnios, performance, smartos, Solaris, usemethod, zones

USENIX LISA 2012: Performance Analysis Methodology

At USENIX LISA 2012, I gave a talk titled Performance Analysis Methodology. This covered ten performance analysis anti-methodologies and methodologies, including the USE Method. I wrote about these in the ACMQ article Thinking Methodically about Performance, which is worth reading for more detail. I’ve also posted USE Method-derived checklists for Solaris- and Linux-based systems. The [...]

Read more...
Posted on December 13, 2012 at 3:51 pm by Brendan Gregg · Permalink · Comments Closed
In: methodology, performance, slides, talk, usemethod