Read/Write Performance Observations

As I mentioned in my previous post, we have now moved more active mailboxes to the Sun Oracle 7000 Storage System. Active means, incoming mails, POP3/IMAP4 accesses etc.

Reads

First we will take a look at disk read latency. We can see that more than 50% are lower than 10ms.







A sanity check on one of the NFS-clients confirms this

    r/s    w/s   kr/s   kw/s wait actv wsvc_t asvc_t  %w  %b device  
   73.6   36.5  494.4  317.7  0.0  0.3    0.1    2.4   1  22 filer:/filesystemb 
   39.8   31.7  322.2  445.4  0.0  0.1    0.4    1.8   1  12 filer:/filesystemb

We can see that the average service time over this 10 second sample is around 2ms. Usually I'm not interpreting the %b value too much, other than if it constantly is at 100%. Newer file-systems read/write in bursts, which makes it not a good problem indicator.

Next we will take a look at the read I/Os. We can see a moderate number of read I/Os per disk.







We can also see the bandwith usage for this operation. For comparison I've marked the same physical disk.







Writes

Writes are handled differently by the Oracle Storage 7000 systems, as syncronous writes are written to the SLOG/SSD devices. Asynchronous writes are collected into the ARC and then flushed to disk. You can see this as long bars in the latency graph.










A different view of this is the I/O per second graph for writes. The yellow area on the graph are the SLOG devices. You can see an almost straight line, whereas the blue area has some peaks.







The last graph shows the same as I/O bytes per second graph. Here we see about the same peaks.








Read/Write-Ratio

Looking at the read/write-ratio, shows us, that we have much more write I/Os than read I/Os.

 




Summary

Analytics helps us tremendously in monitoring how the system behaves during migration. This is especially important, with workloads, that can not be easily simulated.

As we already had ZFS experience, the patterns have not been too surprising. What comes to me as a surprise is the high write portion. On our FC-attached storage array, we see a ~ 50%/50% read/write ratio for the same workload using VxFS. While not being a problem at all, I will investigate the reason for this further.

Comments

Post a Comment

Popular posts from this blog

Heating up the Data Pipeline (Part 1)

Image
Pre-Processing Data I often hear the question from our customers, how data can be transformed prior to indexing in Splunk. Damien from  Baboonbones  has done a tremendous job in creating add-ons   providing custom inputs for Splunk. Most of his custom inputs provide the means to pre-process data by allowing custom event handlers to be written. Sometimes you still want to pre-process data that gets collected from Splunk's standard input types, like file monitors, Windows EventLogs, scripted inputs etc. Also, not everyone is capable of writing custom event handlers. A requirement these customers have, is that they have rolled out a large number of Splunk Universal Forwarders and they do not want to install another agent. To summarize, the solution capable of pre-processing data, should be easy to use, be easily integrated and be build on top of their existing architecture. How to plumb Splunk Pipelines Splunk has its own fittings to connect a Universal Forwar
Read more

Heating up the Data Pipeline (Part 3)

Image
Welcome back to the "Heating up the Data Pipeline" blog series. In part 1 we talked about how to route data from Splunk to a 3rd party system. In part 2 walked through a simple data flow that passes data collected from Splunk Forwarders through Apache NiFi back to Splunk over the HTTP Event Collector. In this part, we will look at a more complex use case, where we route events to an index, based on the sending host's classification. The classification will be looked up from a CSV file. We will make use of Apache NiFi's new Record-Oriented  data handling capabilities, which will look initially a bit more complicated, but once you grasp it, it will make further Use Cases easier and faster to build. High-Level Dataflow We will again start with our ListentTCP input, but this time we will send the data to another Processor Group. The Processor Group will again emit events suitable for Splunk HEC. Note that we have increased the Max Batch Size from 1
Read more

Heating up the Data Pipeline (Part 2)

Image
In Part 1 we went through how to route events from Splunk to a 3rd party system without losing metadata. Now I'll show you how events can be transformed using Apache NiFi and be sent back to Splunk into the  HTTP Event Collector . Note: The following is not a step-by-step documentation. To learn how to use Apache NiFi you should read the G etting Started Guide . Simple Pass-Through Flow As a first exercise we will create a simple flow, that only passes data through NiFi, without applying any complex transformations. The following picture shows a high-level NiFi flow, that receives events in our custom uncooked event format with a TCP listener, then sends the data further into a transformation "black box" (aka Processor Group), which emits events in a format, that can be ingested into a Splunk HTTP Event Collector input. Apache Nifi currently provides a rapidly growing number of processors (currently 266), which can be used for data ingestion, transfo
Read more