your rating and comments does matter
T10 is a Sub-committee of INCITS committee that is an ANSI accredited standards Development Organization.
ANSI T10 standard provide a way to check the integrity of read and write from the host bus adapter (HBA) to the disk and back through SAN Fabric.
DIF (Data Integrity Field)
Data availability in storage is frequently measured simply in terms of reliability of the hardware components and the effect of redundant Hardware.
Also the reliability of storage Software, which is the ability to detect errors, and the ability to correctly report and apply corrective actions to failure have a significant bearing on the overall storage system availability.
T10 standard provide a way to check the integrity of read and write from the host bus adapter (HBA) to the disk and back through SAN Fabric. The DIF includes extending the disk sector from its traditional 512 bytes to 520 bytes, by adding eight additional protection bytes. The extended sector is defined for Small Computer System Interface (SCSI) devices.
In type 1 implementation of T10, the 8Bytes DIF consists of a ReF tag (2bytes), an APP tag (2bytes), and a CRC (2bytes).
Normally on write, the DIF is generated by the HBA, based on the block data and the logical block address. The DIF field is added to the end of the data block, and the data is sent through the fabric to the storage target. The storage target validates the CRC and ReF tag. If correct, stores both the data block and the DIF on physical media, If the CRC doesn’t match the data, then the data was corrupted during the write and the write operation is returned back to the host with write error code.
On Read, the DIF is returned along with the data block to the host, which validates the CRC and ReF tag. the validation adds a very small amount of latency (milliseconds IO) which may impact the overall response on a small block transaction <4K
T10 Protection Information (PI)
allows the checksum to be transmitted from the HBA and the application to the disk drive.
Traditionally, protecting the Integrity of customer’s data was done using ECC, and CRC, but this has been a gap across the IO path from the OS to the storage.
T10 PI ensures the data is validated as it moves through the data path enabling seamless end-to-end integrity.
T10 and 3PAR
3PAR StoreServ OS offers automated end-toend error checking during data frames journey through 3PAR StoreServ Array to the disk devices to help ensure data integrity in support of Tier-1 resiliency.
In 3PAR StoreServ Systems, DIF check is available through:
- CRC/Parity checks on all internal CPUs.
- Control Cache ECC checks.
- PCIe I2C CRC/Parity checks.
- Data Cache ECC checks.
- 3PAR ASIC connection CRC/Parity checks.
- protocol (FCoE, iSCSI, FC) CRC checks at the frame level (accelerated via ASICs)
- Disk devices CRC checks at the block level, which occur once data has landed and through its life cycle.
CRC error checking has also been extended to replicated data with remote copy, which helps ensure potential cascade data issues don not occur. Also StoreServ replication includes a link pre-integration test to verify the stability of remote copy replication links in advance.
All 3PAR StoreServ 20000 and 7000 Systems are formatted with 520 byte blocks in order to provide space to store the CRC logical block guard (T10 DIF for each block. This value is computed by HBA before writing a block and is checked when block is read.
SATA doesn’t support 520-byte blocks.
AT the end, StoreServ continuously runs a background “PD Scrubber” process to scan all blocks of the physical drives in the system, which helps to proactively detect and correct any low level CRC and but errors.