Performance and Scalability track

Friday, August 21, 2015 from 9:0011:30am
Metropolitan A

The Linux Plumbers 2015 microconference – Performance and Scalability track focuses on scalability, both upwards and downwards as well as up and down the stack. There are quite a few active projects and efforts of a wide range that aim at enhanncing performance and scalability both in the Linux kernel and in user-space projects. In fact, one of the purposes of this forum is for developers from different projects to meet and collaborate – both kernel developers and researchers, doing more experimental work. After all, for the user to see good performance and scalability, all relevant projects must perform and scale well.

Because performance and scalability are very generic topics, this track is aimed at issues that are not addressed in other, more specific sessions, such as networking.

The structure will be similar to what was followed the previous years (2011, 2012 and 2013): about 30 minutes per subject with discussion.

Microconference Leader

Davidlohr Bueso

Proposals for this track

* Do Virtual Machines Really Scale?

Scalability issue of VMs on 80core and 120core machines.
Performance and Scalability 06/08/2015
Sanidhya Kashyap

* Fast Bounded-Concurrency Hash Tables

This talk introduces a general technique for achieving single-writer non-blocking hash tables at low to negligible cost.
Performance and Scalability 06/23/2015
Samy Bahra

* How to apply mutation testing to the RCU for fun and profit

Learn how we overcame the practical and computational limitations of mutation testing to identify real problems in projects as complex as the Kernel, and how you too can adopt this technique.
Performance and Scalability 06/10/2015
Iftekhar Ahmed

* Improving Transactional Memory Performance with Queued Locking

We present a new modified queued locking method that improves transactional memory technology (like Intel TSX), by using queued locking to adaptively control the number of threads doing speculative transaction for optimal performance. (slides)
Performance and Scalability 06/16/2015
Tim Chen

* Limitations of various linkers in clang based toolchains

Limitations of various linkers in clang based toolchains (slides)
Performance and Scalability 06/30/2015
Mark Charlebois

* Pimping Futex Performance

A look at the past, present and future of futex performance.
Performance and Scalability 06/17/2015
Davidlohr Bueso

* Standardization and RCU

Report on how RCU is faring in the C/C++ standards committees
Performance and Scalability 06/16/2015
Paul McKenney

* Unfair Queued Spinlocks and Transactional Locks

A performance review on unfair queued spinlocks for virtual guests and transactional locks based on the Intel's TSX technology. (slides)
Performance and Scalability 07/03/2015
Wai Man Long