170 IBM System Storage N series Hardware Guide Improves response times. Both block-oriented SAN protocols (Fibre Channel protocol,iSCSI, and FCoE) and file-oriented NAS storage protocols (CIFS and NFS) require anacknowledgement from the storage system that a write was completed. To reply to a writerequest, a storage system without any NVRAM must complete the following steps:a. Update its in-memory data structures.b. Allocate disk space for new data.c. Wait for all modified data to reach disk.A storage system with an NVRAM write cache runs the same steps, but copies modifieddata into NVRAM instead of waiting for disk writes. Data ONTAP can reply to a writerequest much more quickly because it must update only its in-memory data structures andlog the request. It does not have to allocate disk space for new data or copy modified dataand metadata to NVRAM. Optimizes disk writes. Journaling all write data immediately and acknowledging the clientor host not only improves response times, but gives Data ONTAP more time to scheduleand optimize disk writes. Storage systems that cache writes in the disk driver layer mustaccelerate processing in all the intervening layers to provide a quick response to host orclient. This requirement gives them less time to optimize.For more information about how Data ONTAP benefits from NVRAM, see IBM SystemStorage N series File System Design for an NFS File Server, REDP-4086, which is availableat this website:http://www.redbooks.ibm.com/abstracts/redp4086.html?Open11.4.2 NVRAM operationNo matter how large a write cache is or how it is used, eventually data must be written to disk.Data ONTAP divides its NVRAM into two separate buffers. When one buffer is full, thattriggers disk write activity to flush all the cached writes to disk and create a consistency point.Meanwhile, the second buffer continues to collect incoming writes until it is full, and then theprocess reverts to the first buffer. This approach to caching writes in combination with WAFLis closely integrated with N series RAID 4 and RAID-DP. It allows the N series to schedulewrites such that disk write performance is optimized for the underlying RAID array. Thecombination of N series NVRAM and WAFL in effect turns a set of random writes intosequential writes.The controller contains a special chunk of RAM called NVRAM. It is non-volatile because ithas a battery. Therefore, if a sudden disaster that interrupts the power supply strikes thesystem, the data that is stored in NVRAM is not lost.After data gets to an N series storage system, it is treated in the same way whether it camethrough a SAN or NAS connection. As I/O requests come into the system, they first go toRAM. The RAM on an N series system is used as in any other system; it is where DataONTAP does active processing. As the write requests come in, the operating system alsologs them in to NVRAM.NVRAM is logically divided into two halves so that as one half is emptying out, the incomingrequests fill up the other half. As soon as WAFL fills up one half of NVRAM, WAFL forces aconsistency point (CP) to happen. It then writes the contents of that half of NVRAM to thestorage media. A fully loaded system does back-to-back CPs, so it is filling and refilling bothhalves of the NVRAM.Upon receipt from the host, WAFL logs writes in NVRAM and immediately sends anacknowledgment (ACK) back to the host. At that point from the host’s perspective, the datawas written to storage. But in fact, the data might be temporarily held in NVRAM.
