Cheetah 10K.6 FC Product Manual, Rev. B

11

The buffer can be divided into logical segments (using Mode Select Page 08h, byte 13) from which data is read and to which data is written.

The drive keeps track of the logical block addresses of the data stored in each segment of the buffer. If the cache is enabled (see RCD bit in the Fibre Channel Interface Manual), data requested by the host with a read command is retrieved from the buffer, if possible, before any disc access is initiated. If cache operation is not enabled, the buffer (still segmented with the required number of segments) is still used, but only as circular buffer segments during disc medium read operations (disregarding Prefetch operation for the moment). That is, the drive does not check in the buffer segments for the requested read data, but goes directly to the medium to retrieve it. The retrieved data merely passes through some buffer segment on the way to the host. On a cache miss, all data transfers to the host are in accordance with buffer-full ratio rules. On a cache hit, the drive ignores the buffer-full ratio rules. See the explanation provided with the information about Mode Page 02h (disconnect/ reconnect control) in the Fibre Channel Interface Manual.

The following is a simplified description of the prefetch/cache operation:

Case A—read command is received and the first logical block is already in the cache:

1.Drive transfers to the initiator the first logical block requested plus all subsequent contiguous logical blocks that are already in the cache. This data may be in multiple segments.

2.When a requested logical block is reached that is not in any cache segment, the drive fetches it and any remaining requested logical block addresses from the disc and puts them in a segment of the cache. The drive transfers the remaining requested logical blocks from the cache to the host in accordance with the Mode Select Disconnect/Reconnect parameters, page 02h.

3.If the prefetch feature is enabled, refer to section 4.5.2 for operation from this point.

Case B—A Read command requests data, and the first logical block is not in any segment of the cache:

1.The drive fetches the requested logical blocks from the disc and transfers them into a segment, and then from there to the host in accordance with the Mode Select Disconnect/Reconnect parameters, page 02h.

2.If the prefetch feature is enabled, refer to section 4.5.2 for operation from this point.

During a prefetch, the drive crosses a cylinder boundary to fetch data only if the Discontinuity (DISC) bit is set to 1 in bit 4 of byte 2 of the Mode Select parameters page 08h. Default is zero for bit 4.

Each cache segment is actually a self-contained circular buffer whose length is an integer number of logical blocks. The wrap-around capability of the individual segments greatly enhances the cache’s overall perfor- mance, allowing a wide range of user-selectable configurations. The drive supports operation of any integer number of segments from 1 to 32. Divide the 6,877 Kbytes in the buffer by the number of segments to get the segment size in bytes; then divide by the sector size to get the number of sectors per segment, any partial sec- tors remaining are not used. Default is 3 segments.

Note. The size of each segment is not reported by Mode Sense command page 08h, bytes 14 and 15. The value 0XFFFF is always reported regardless of the actual size of the segment. Sending a size specifi- cation using the Mode Select command (bytes 14 and 15) does not set up a new segment size. If the STRICT bit in Mode page 00h (byte 2, bit 1) is set to one, the drive responds as it does for any attempt to change an unchangeable parameter.

4.5.1Caching write data

Write caching is a write operation by the drive that makes use of a drive buffer storage area where the data to be written to the medium is stored while the drive performs the Write command.

If read caching is enabled (RCD=0), then data written to the medium is retained in the cache to be made avail- able for future read cache hits. The same buffer space and segmentation is used as set up for read functions. The buffer segmentation scheme is set up or changed independently, having nothing to do with the state of RCD. When a write command is issued, if RCD=0, the cache is first checked to see if any logical blocks that are to be written are already stored in the cache from a previous read or write command. If there are, the respective cache segments are cleared. The new data is cached for subsequent Read commands.

If the number of write data logical blocks exceed the size of the segment being written into, when the end of the segment is reached, the data is written into the beginning of the same cache segment, overwriting the data that

Page 21
Image 21
Seagate ST3146807FC, ST336607FC, ST373307FC manual Caching write data

ST336607FC, ST373307FC, ST3146807FC specifications

The Seagate ST373307FC, ST3146807FC, and ST336607FC are robust hard disk drives designed for enterprise environments where reliability and performance are critical. Each model offers distinct features suited to various data storage needs, reflecting Seagate's commitment to innovation in storage technology.

The ST373307FC is a 73GB Fibre Channel hard drive that operates at 15,000 RPM, making it exceptionally fast for read and write operations. This drive employs a 4GB cache, enhancing its performance by allowing quicker access to frequently used data. With a 3.5-inch form factor, the ST373307FC is built for high-density environments, delivering reliable access to data. Its Fibre Channel interface supports high-speed connections, essential for data-intensive applications like databases and virtualization.

Next in line, the ST3146807FC offers a capacity of 146GB, also operating at 15,000 RPM. This drive features an 8GB cache, which significantly improves its data transfer rates, making it ideal for transactional workloads. The ST3146807FC is engineered to minimize latency and maximize throughput, ensuring that enterprises can access critical data swiftly and efficiently. Additionally, its rugged construction is designed to withstand the demands of continuous operation, providing peace of mind for businesses with strict uptime requirements.

Lastly, the ST336607FC has a substantial storage capacity of 300GB and shares the 15,000 RPM operational speed that characterizes this high-performance line. With a similar 8GB cache as the ST3146807FC, it offers excellent performance for read-intensive and write-heavy applications. The ST336607FC is also designed with enhanced thermal management features, supporting cooler operations in densely packed server environments. This model is particularly suited for applications such as data warehousing and online transaction processing, where speed and reliability are paramount.

Each of these drives utilizes advanced technologies like Native Command Queuing (NCQ), which optimizes the order of read and write requests. These features, combined with Seagate's reputation for durability and performance, make the ST373307FC, ST3146807FC, and ST336607FC excellent choices for enterprise applications requiring high-speed data access and exceptional reliability. In environments where data integrity and uptime are critical, these drives provide the robust performance necessary to meet demanding storage needs.