System News for Sun Users - The Blog

DTrace Aids in Identifying Latency Causes

A duel in the Suns of sorts involving a Sun Fire V40z and a Sun Fire X4450 connected back-to-back was performed to discover their ability to exchange UDP or TCP messages with a round-trip latency under 100 ms per message. In addition, both servers were occupied with some business logic, to bring the overall CPU utilization to 60-80% on each machine. Both were running Solaris 10 Update 5 and Java RTS 2.1 (build 1.5.0_16_Java-RTS-2.1_fcs-b11_RTSJ-1.0.2). Frederic Pariente reports the results in his blog “Real Time opens system and network surveillance to Java,” a proof-of-concept to evaluate Java RTS 2.1 on Solaris 10.

(Get More Information . .)

Why a Million IOPS Capable Platform is Relevant Even Without a Million IOPS Demand

The Sun Oracle Database Machine is a million IOPS-capable platform. Why would a CIO consider it as a solution when a particular datacenter may not require so much I/O? Performance Architect in Oracle’s Systems Technology Group Kevin Closson gives reasons and applies these to what really goes on in real world production datacenters, touching on chaos, IOPS, latency and Flash.

(Get More Information . .)

Sun & Thomson Reuters Joint Engineering Offers Increased Capacity, Improved Efficiencies, Reduced Latency

Solaris Containers on next-generation multi-core platforms is giving customers using Reuters Market Data Systems (RMDS) in Solaris environments increased capacity, improved efficiencies, reduced latency, and overall enhanced performance while also reducing total cost of ownership. Sun and Thomson Reuters claim recent benchmarks have demonstrated that RMDS performs better in a virtualized environment with Solaris Containers, which delivers strong performance improvements across a significantly reduced hardware footprint, than one running RMDS on a number of individual Sun server machines.

(Get More Information . .)

Sun Engineer Brendan Gregg on the Fishworks team discovered that vocal vibrations affect disk latency through the use of Analytics from Fishworks. In a two-minute YouTube video that has garnered the attention of over a quarter of a million people, Gregg demonstrates how his vocal vibrations cause a sharp spike in the number of I/O operations per disk and a noticeable latency increase on the overall workload.

He offers a screenshot of Analytics on a Sun Storage 7410 Unified Storage System. He measures disk I/O operations broken down by latency and disk I/O operations that take at least 520 ms broken down by disk using DTrace for performance analysis of the disks. He applies a right workload to two JBODs and then proceeds to yell into them.

The effect of disk vibration is clearly evident. The screenshot shows the latency of the disks has risen and the specific disks affected demonstrate I/O operations longer than 520 ms.

“Amazing stuff,” Gregg pronounces. “This has been made possible by Analytics from Fishworks, which lets us look at things we’ve never seen before.”