High-end backup compression numbers

On Tuesday we had a look around our customer's data centers here in Austria – hidden away in the labryinthine bowels of a very large building in Vienna. Typical data centre with a halon fire extinguishing system but exceptional in its neatness and organization. The star of the show was their new HP Superdome – 32 dual-core Itaniums with 1/2 a terabyte of memory. Very nice – providing a lot of headroom for their workload to grow (currently at four hundred thousand SQL statements per *second* – spelled out to show there's no accidentally added zeros).

One of the systems they showed us is responsible for doing backups. They stripe the backups across 12 devices using multiple network cards and can manage to backup 2 terabytes in two hours after tweaking the BLOCKSIZE, BUFFERCOUNT, and MAXTRANSFERSIZE! Now comes the cool(er) part – with backup compression on SQL Server 2008 they've benchmarked backing up 2 terabytes in 36 minutes! That's a pretty awesome number and makes for some excellent disaster recovery times.

(Details printed with permission)

4 thoughts on “High-end backup compression numbers

  1. Hello Paul,

    I have a little question about the backup devices :

    were the 12 devices rely on separated physical disks in the SAN (for example, 2 physical disks = 1 logical volume = 1 device ) or was it only 12 volumes created in the SAN using stripping technique or similar, using shared disks (example, 24 physical disk = 12 volumes = 12 devices, using stripping)

    I hope it is clear ……

    Keep on the good work and thank you !

  2. Thats awesome, i’d love to have access to one of those superdome servers. Just a question, what OS and which SQL implementation is it running.

Leave a Reply

Your email address will not be published. Required fields are marked *

Other articles

Imagine feeling confident enough to handle whatever your database throws at you.

With training and consulting from SQLskills, you’ll be able to solve big problems, elevate your team’s capacity, and take control of your data career.