Category Archives: working

working

Statistics & Visualizations: Two Lightning Talks by Brendan Gregg

At USENIX LISA 13, Brendan Gregg led a full-day, interactive workshop on metrics, with attendees from across the tech industry. Tutorial material was provided by Brendan, along with Caskey Dickson of Google and Theo Schlossnagle of OmniTI. There were also group exercises and presentations. I filmed it all, and we’ll be rolling it out to everyone soon.

At the end of the day, Brendan gave two quick summary talks on statistics and visualizations, presented here as a teaser.

The full playlist from the workshop is here.

OpenZFS Developer Summit

Spent the day filming and streaming this. The full playlist is here, below are links to each individual talk.

Systems Performance Book Videos

While Brendan was writing his new book Systems Performance: Enterprise and the Cloud, we filmed some short videos about why he wanted to write a systems performance book, how hard it was to get started, and about a few chapters as he completed them…

Then life became busy and insane (partly because he was writing a book, but not only), and we didn’t shoot anything at all for months. And then, finally, the book was finished.

Many months after that, when the book was finally in print, we launched it at a meetup of BayLISA, where Brendan gave a long talk on all of it, with in-depth look at Chapter 6: CPUs.

The book is also available from Informitin paper, ebook, or a bundle of both!

Marketing Your Tech Talent

I was thrilled to have the opportunity to present this talk at the Monktoberfest in Portland, Maine.

Slide deck with notes – Marketing Your Tech Talent

Some very kind words about the talk from Joe Brockmeier here.

I’m just gonna brag a little here:

“Datashits” was a Freudian slip. Really. I promise.

Monktoberfest tweets
Screen shot 2013-10-04 at 5.07.11 PM
Monktoberfest tweets
Screen shot 2013-10-04 at 5.32.56 PM
Screen shot 2013-10-04 at 5.06.50 PM
Screen shot 2013-10-04 at 5.06.10 PM

Why 4K?

George Wilson at ZFS Day

For over 30 years, hard drives have designated the smallest storage location as 512 bytes. In January 2011, all major hard drive manufactures began shipping their hard drive platforms using a new standard called Advanced Format. To aid in the transition, these new hard drives provide a 512 byte emulation mode that allows the drives to advertise themselves as a 512 byte addressable devices. This can severely impact write performance resulting in the need for read-modify-write operations for any misaligned or partial writes that are issued.

The problem is not limited to just physical hardware. Other storage platforms may also provide LUNs (logical unit number) that presents themselves as a 512 byte addressable devices when, in fact, they use a 4K sector size internally. Although ZFS has built-in support for 4K sectors, it has no automatic way of dealing with the lies that the storage devices tell. This talk will focus on the methods that have been developed to work around the lies that hard drive storage platforms tell and will discuss the challenges and drawbacks that come with using 4K sectors.