I hope this will clear the air on the Magstar E technology. I have seen a
signficant amount of information on the list that is right from the point of
the postee, but not generally factually correct. The answer always is "It
depends".
I get 20+MB/sec on my DataBase backup to the Magstar E Drive. How? The
source is ESS disk that is striped across three different arrays. I am
going to change it to 8 this week to see if I can get it to improve further.
No mirroring is necessary.
Here are the actual IBM published numbers on the Magstar E drive.
40 MB/Sec with 3:1 Compression
14.1 MB/Sec at the head
This is very block dependent. You have to write 128KB blocks to get in this
ballpark.
The new H drive is rated at 70 MB/Sec, 5:1 Compression still 14.1 at the
head.
I have actually seen client backups to a single E drive run at 50 MB/sec on
both SAN and Gigabit.
The bottle neck for the tape drive is usually the source cannot provide the
data or the data is not very compressible. You can tell about the
compression by looking at the estimated capacity number for the tapes in a
pool after they are full. When you are getting 180 GB on a tape you are
getting 4.5:1 compression
Here is an example of a tape that got 200GB on it. This one just so happens
to only now be 66.7 percent full. This had oracle database data on it.
VOLUME_NAME: VSNXX3
STGPOOL_NAME: MY_STGPOOL
DEVCLASS_NAME: DC_ATL-2-3590E1A
EST_CAPACITY_MB: 200666.0
PCT_UTILIZED: 66.7
STATUS: FULL
ACCESS: READWRITE
PCT_RECLAIM: 33.2
SCRATCH: Yes
ERROR_STATE: No
NUM_SIDES: 1
TIMES_MOUNTED: 5
WRITE_PASS: 1
LAST_WRITE_DATE: 2002-06-12 08:23:09.000000
LAST_READ_DATE: 2002-08-25 07:55:35.000000
PENDING_DATE:
WRITE_ERRORS: 0
READ_ERRORS: 0
LOCATION:
CHG_TIME: 2002-06-10 13:28:23.000000
CHG_ADMIN:
This one is even better. A whopping 355GB on one Magstar K tape from an E
drive. Now, that is nearly 9:1 compression.
VOLUME_NAME: VSNXX1
STGPOOL_NAME: CPY_MY_STGPOOL
DEVCLASS_NAME: DC_ATL-2-3590E1A
EST_CAPACITY_MB: 355481.1
PCT_UTILIZED: 80.2
STATUS: FULL
ACCESS: OFFSITE
PCT_RECLAIM: 19.7
SCRATCH: Yes
ERROR_STATE: No
NUM_SIDES: 1
TIMES_MOUNTED: 6
WRITE_PASS: 1
LAST_WRITE_DATE: 2002-06-15 04:26:21.000000
LAST_READ_DATE: 2002-06-10 04:39:06.000000
PENDING_DATE:
WRITE_ERRORS: 0
READ_ERRORS: 0
LOCATION: VAULT
CHG_TIME: 2002-06-17 12:29:14.000000
CHG_ADMIN: CONTROLM
Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180
-----Original Message-----
From: Remco Post [mailto:r.post AT SARA DOT NL]
Sent: Monday, September 02, 2002 9:17 AM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: backup performance with db and log on a SAN
On maandag, september 2, 2002, at 01:49 , Daniel Sparrman wrote:
> Hi Eliza
>
> As I understand it, each "tape-HBA" has 3 3590E FC connnected to it.
> The two 21G database disks, are each connected to it's own HBA?
>
> According to spec sheets, the 3590E FC could handle speed up to
> 100MB/s with 3:1 compression. This means that with 3 drives, you could
> have up to 300MB/s. However, the HBA will only theoretically handle
> 125MB/s, or 250MB/s with 2Gb FC.
>
Fortunately, an 3590E will do only about 16MB/s or so, so 3 drives will not
fill up a FC link.... 3:1 compression is very rare in my experience, 2:1 or
less is far more likely...
---
Met vriendelijke groeten,
Remco Post
SARA - Stichting Academisch Rekencentrum Amsterdam http://www.sara.nl
High Performance Computing Tel. +31 20 592 8008 Fax. +31 20 668 3167
PGP keys at http://home.sara.nl/~remco/keys.asc
"I really didn't foresee the Internet. But then, neither did the computer
industry. Not that that tells us very much of course - the computer industry
didn't even foresee that the century was going to end." -- Douglas Adams
|
