Changes

In order to have our '''Koji ''' Build Farm run as efficiently as possible we needed to find out which form of data storage would be the fastest overall. The candidates were: * '''PATA:''' Hard Drive connected via USB.* '''NFS:''' Share from HongKong.* '''iSCSI:''' Network connection to HongKong. ===What results are we interested in?=== There are 3 main results that we are interested in when rating storage performance.

* PATA Hard Drive connected via USB'''Read:''' The amount of data that can be read from the storage medium per second.* NFS share from HongKong'''Write:''' The amount of data that can be written to the storage medium per second.* iSCSI network connection '''Access:''' Time required for a computer to HongKongprocess data from the processor and then retrieve the required data from a storage medium.

1Since '''NFS''' and '''iSCSI''' are both network storage solutions they have no cost in themselves, but rely on network storage on a remote server. Read: The amount This price is simply the cost of data that can be read from the storage medium per second.<br>2. Write: The amount of data drives that can will be written to installed in the remote storage medium per secondserver.<br>3. Access: Time required for A USB connected PATA or SATA drive requires both a computer to process data from the processor hard drive and then retrieve the required data from a storage medium'''PATA/SATA''' to '''USB''' interface such as an external drive enclosure.

Another factor is cost. Since * '''NFS and :''' Free (Uses existing storage)* '''iSCSI are both network :''' Free (Uses existing storage solutions they have )* '''USB PATA:''' ~$100 CAD ===Pictures===http://david-chisholm.no cost in themselves, but rely on network storage on a remote server. This price is simply the cost of the drives that will be installed in the remote storage server-ip. A USB connected PATA or SATA drive requires both a hard drive and a PATAorg/SATA to USB interface such as an external drive enclosurenetworkdiagram.jpg

*Benchmark using a linux untiliy called '''Bonnie++''' written by Russell Coker.*The Benchmark was run 3 times on each medium, the results were then averaged together. *The command used is as follows:

===What happened while you worked on was the problem? You had multiple iterations -- what happened at each milestone? Did you go down the wrong path and have process we used to start over? What barriers did you encounterchoose our benchmarking solution?The process was simple, find a storage solution that would result in the best build times while using the most efficient use of the storage resources available to us.===

The main issue encountered goal was finding to find a repeatable benchmarking storage solution what that would give result in the desired results best build times while being able to test all 3 using the most efficient use of our the storage mediums. Common Linux tools such as the DD command are capable of doing disk benchmarking, but will only work of real devices and not network file systemsresources available to us.

The main issue encountered was finding a repeatable benchmarking solution what would give the desired results while being able to test all 3 of our storage mediums. Common Linux tools such as the '''DD''' and '''HDPARM''' commands are capable of doing disk benchmarking, but will only work for physical devices and not network networked ones, making them useless tests for our purposes. The solution was '''Bonnie++''', a Linux command line utility which gives an extensive amount amount of storage performance information while also having the ability to test all of our storage systems. ===Pictures=== http://www.business-strategy-innovation.com/uploaded_images/Innovation-Process-799858.jpg

We discovered that finding a viable benchmarking solution is harder then it sounds. Raw data will not always correspond with real results as it comes down to ===What did we discover during the application using those resources.process?===

We discovered that finding a viable benchmarking solution is harder then it sounds. Raw data will not always correspond with real results as it comes down to the application using those resources. This is evident in the mock tests using '''NFS''' vs '''USB PATA''' where '''USB PATA''' performed faster even though its benchmark results were lower using '''Bonnie++'''. ===Pictures=== http://thumbs.dreamstime.com/thumb_122/1171633495w0G6A0.jpg =Issues= ===USB PATA===*Works without issue. ===NFS===*Works, but results in longer build times than USB PATA even though it benchmarked at higher speeds. ===iSCSI===*Seems to work at first, but only to a point.*We can login to an initiator, however, under heavy load the target receives invalid opcodes, causing the connection to fail.*Experimenting with a alignment value of 3 did not clear the issue.*Using the exact same target with a F12 x86_64 initiator is successful, issue seems to be '''ARM''' related. ===Pictures=== http://exportabel.files.wordpress.com/2009/10/train_wreck_at_montparnasse_1895.jpg =Bonnie++ Results=

<table border="1" cellspacing="01" width="50%0">

<table border="1" cellspacing="01" width="500%">

<table border="1" cellspacing="01" width="500%">