storage ICT
Figure 2 – Class of Service Alert with Emulex OneCommand™ Vision
Consider Figure 1 – here we see the traditional analysis of an application slowdown or outage. Is it server or storage related? If you have ever been in this situation, you know that sometimes it can take days to figure out what is going on. Meanwhile, you have your boss breathing down your neck and this pressure just works its way up the management chain the more critical the application affected is and the time it takes to figure out what’s going on. With OneCommand™ Vision, in five minutes, you can rule out that you have a storage issue.
Following the diagram’s flow, number 1 shows the dashboard feature of OneCommand™ Vision. This is the “tremendous” part. In one screen, you can see a convenient summary view of potential performance issues. The dashboard features several reports that highlight resources potentially experiencing resource performance or availability issues.
Next in number 2, we get to the innovative part. Using the Analytics Correlation Engine (ACE), the I/O Profile module presents important information derived from I/O transactions and storage infrastructure configuration. It also provides historical roll-ups of multiple metrics, including total number of I/O, average number of initiators per ports and “top talkers,” allowing you to quickly optimize existing storage infrastructure, resulting in a more load-balanced network.
In number 3, we see the sophisticated side of OneCommand™ Vision, the I/O Analyzer, which allows you to quickly resolve, and even predict, complex data centre performance and availability issues by providing deep SCSI and wire protocol analysis that highlights hidden system issues that normally go undetected. OneCommand™ Vision proactively collects and analyses important protocol events, such as busy and queue full errors from the storage transport layer, enabling you to quickly resolve and prevent the most complex data centre outages.
Which brings us to number 4, the easy part – in just five minutes, OneCommand™ Vision has determined that storage is not an issue! You don’t waste any cycles chasing a problem that doesn’t exist. You go straight to problem solving on the server side (in this example).
What Else Can OneCommand™ Vision Do? How would you like to set your I/O performance levels and be alerted when certain thresholds occur? Yes – with OneCommand™ Vision’s cutting edge monitoring of LUN-based service level agreements (SLAs). Figures 2 and 3 provide a few more examples of how OneCommand™ Vision can save hours and hours of time across your data centre.
In Figure 2, you can see how to automatically monitor a class of service delivered by storage to applications where you set the service based on actual usage data. You configure throughput (MB/sec) and workload (IOPS) thresholds up front, and then OneCommand™ Vision sends you an alert when that threshold is reached.
Do you ever have a situation where you want to understand if there was a momentary spike in storage I/O performance, without the expense and complexity of a network analyser? OneCommand™ Vision uses a histogram technique to pick up instantaneous spikes in I/O response times that can be correlated to the time frame of the slow-down in question.
In Figure 3, you can see how to automatically ensure storage performance expectations across a range of I/O sizes, with flexible management of complex application latency expectations. Here you configure I/O size range (kb), expected I/O latency (ms) thresholds up front, and then OneCommand™ Vision sends you an alert when that threshold is reached.
Figure 3 – Service Performance Alert with Emulex OneCommand™ Vision June 2012 I
www.dcseurope.info 27
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56