Solid State disk for a desktop system C drive?

On 19 Jan 2009 21:02:28 GMT, Arno Wagner put finger
to keyboard and composed:

Previously Franc Zabkar wrote:
On 19 Jan 2009 13:58:53 GMT, Arno Wagner put finger
to keyboard and composed:

Previously Franc Zabkar wrote:
On Fri, 16 Jan 2009 18:51:01 -0600, Chris put finger
to keyboard and composed:

You may have different priorities, but I have only the SSD drive in that
system and enjoy the silence and the low heat inside the box.

Anyway, have fun and if you don't mind share your experience when you have
time.

Speaking of sharing, I'm curious whether SMART is implemented in these
drives, and how? Are there any SMART extensions that are peculiar to
SSDs?

My OCZ does only give temperature and vendor stuff. Goven that
most HDD measures (including reallocated sectors) do not make sense,

How would bad areas of memory be accounted for?

The problem here is that SSDs do not use the 512 byte sector
size (instead something much, much larger, like 128kB) and
do not use the HD reallocation mechanism. I would like to
see something like "failed reads" or "ECC recoverd bytes", and
it is possible that the drive actually states them, but in vendor
specific attributes that at least smartctl cannot interpret.
If anybody knows more, below is a slightly shortened SMART
dump of my 30GB OCZ SSD.

Arno


=== START OF INFORMATION SECTION ===
Device Model: OCZ CORE_SSD
Firmware Version: 02.10104
User Capacity: 32,044,482,560 bytes
Device is: Not in smartctl database [for details use: -P showall]
ATA Version is: 7
ATA Standard is: ATA/ATAPI-7 T13 1532D revision 4a
Local Time is: Mon Jan 19 21:59:15 2009 CET
SMART support is: Available - device has SMART capability.
SMART support is: Enabled

SMART Attributes Data Structure revision number: 1280
Vendor Specific SMART Attributes with Thresholds:
ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE
12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 241
9 Power_On_Hours 0x0032 100 100 000 Old_age Always - 0
194 Temperature_Celsius 0x0007 032 100 000 Pre-fail Always - 0
229 Unknown_Attribute 0x0002 100 000 000 Old_age Always In_the_past 259986015180268
232 Unknown_Attribute 0x0002 100 048 000 Old_age Always - 9028846498104
233 Unknown_Attribute 0x0002 100 000 000 Old_age Always In_the_past 0
234 Unknown_Attribute 0x0002 100 000 000 Old_age Always In_the_past 592722311424
235 Unknown_Attribute 0x0002 100 000 000 Old_age Always In_the_past 187423

I would think that SSDs would use wear levelling techniques and that
important indicators of old age would be the number of times that each
memory block had been rewritten. Perhaps one of the attributes
reflects this.

I'd monitor the SMART data on a daily basis and see whether the
attributes change by incrementing the whole word or particular bytes.
Attribute 229 uses all 48 bits, so it appears that it reflects
multiple parameters. OTOH, attribute 235 probably reflects only one
parameter.

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.