Error detection/correction and fault detection/recovery – Data processing system error or fault handling – Reliability and availability
Reexamination Certificate
1999-12-15
2003-05-27
Beausoliel, Robert (Department: 2785)
Error detection/correction and fault detection/recovery
Data processing system error or fault handling
Reliability and availability
Reexamination Certificate
active
06571354
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to the field of mass storage systems. In particular, the invention relates to the management of storage transactions in, and the configuration of, intelligent storage area networks for the purposes of allocating and changing the allocation of storage resources.
2. Description of the Related Art
The storage of large amounts of data in so-called mass storage systems is becoming a common practice. Mass storage systems typically include storage devices coupled to file servers on data networks. Users in the network communicate with the file servers for access to the data. The file servers are typically connected to specific storage devices via data channels. The data channels are usually implemented with point-to-point communication protocols designed for managing storage transactions.
As the amount of storage increases, and the number of file servers in communication networks grows, the concept of a storage area network (SAN) has arisen. Storage area networks connect a number of mass storage systems in a communication network which is optimized for storage transactions. For example, Fibre Channel arbitrated loop (FC-AL) networks are being implemented as SANs. The SANs support many point-to-point communication sessions between users of the storage systems and the specific storage systems on the SAN.
File servers and other users of the storage systems are configured to communicate with specific storage media. As the storage systems expand or media is replaced in the system, re-configuration is required at the file servers and other users. Also, if a need arises to move a data set from one device to another, in a so-called data migration operation, it is often necessary to block access to the data set during the migration process. After migration is complete, re-configuration at the user system must be executed to make the data available from the new device. The blocking of access to the data set during the transfer is a particularly costly action for large data sets having many users. Both the amount of time required to move a copy of the data to the new device, and the number of people inconvenienced can be very large. The above identified related application entitled Method And System For Managing Data Migration For a Storage System, describes solutions to many of the problems associated with migrating data sets among devices in a storage network.
Also, failures of devices in the storage network can occur. Upon failure of a device in an array, data is lost or performance suffers while the data on the failed device is reconstructed. When failure occurs, replacement devices may be needed to recover network performance. The insertion of replacement devices requires data migration operations from backup systems, or from redundant storage in the network. Thus, device failures cause additional problems for network administration.
Data sets are stored in sets that include arrays of storage devices in order to improve the performance of data storage transactions, and to improve fault tolerance in data storage systems. Common configurations for arrays of storage devices are known as RAID levels. For example, RAID 0 consists of a striped Disk Array. The data in a RAID 0 array is broken down into data sets referred to as blocks, and each block is written on a separate disk drive or storage device. RAID 1 consists of mirrored and duplexed sets of storage devices. RAID 3 consists of a set of storage devices on which data blocks are subdivided into stripes, which are written on multiple storage devices. In addition, stripe parity is generated on writes and stored within the array for each striped data block, and checked during reads of the data. In a RAID 5 arrays, data blocks are written on the disks within the array, and parity for the blocks of the same rank is generated on writes. The block parity is recorded in distributed locations within array, and checked during reads. A variety of other RAID levels are well-known. Recovery from failures of storage devices involved in RAID configurations, or in other sets of storage arrays used to store a data set, involves a variety of mechanisms and procedures which can make administration of a storage system complex.
Modern storage devices, such as hard disk drives, are extremely reliable, with a typical mean time between failure rating of 300,000 hours or more. However, as the number of disk drives per system increases with storage area network technology, and the size of the typical disk drive grows, administrators will experience failures of even very reliable devices. Thus, technology is being developed to elevate the protection of user data. For example, systems have been designed for self-monitoring analysis and reporting in disk drives. For example, the so-called S.M.A.R.T. system developed by Compaq Computer provides for disk drives and other storage devices to generate signals that communicate their predicted reliability status to users and system administrators. With this information, an administrator is able to prevent system downtime, productivity loss and even the loss of valuable data if appropriate corrective action is taken. Other utilities have also been developed for the purposes of diagnosing storage device reliability status.
Overall, as the complexity and size of storage systems and networks increase, the problems of managing failed or worn out storage devices along with configuration of the users of the data and of the storage systems themselves multiply. Accordingly, there is a need for systems that simplify the management of storage systems, and in particular the management of data in devices that need to be replaced, while taking advantage of the flexibility and power of the SAN architecture.
SUMMARY OF THE INVENTION
The present invention provides a method and an apparatus for use in a storage network that facilitates the protection of data in, and replacement of, storage devices that are about to fail before the failure happens. In a network that includes a plurality of sets of storage devices which store respective data sets, a storage device about to fail in one set can be replaced by another storage device from another set of storage devices which is being used to store data having a lower priority. In this manner, the integrity of the higher priority data is maintained, and storage devices that are about to fail are migrated into lower priority storage device sets.
The method comprises assigning priorities to sets of storage devices which store respective data sets in the network. In addition, the method includes detecting a condition of a first particular storage device in a particular set of storage devices that has a first priority. According to various embodiments, conditions which are detected are those which indicate that the first particular storage device is suffering events which make it likely to fail, or otherwise suffering from reduced performance. The conditions are detected for example, by the receipt of a signal from the storage device itself, or by the monitoring of statistics concerning the performance of the storage device. The method of the present invention further provides for selecting a second particular storage device in a second particular set of storage devices having a second priority, which can be used in place of the first particular storage device. In response to detecting the condition, the data set stored in the first particular storage device is migrated to the second particular storage device, and the second particular storage device is identified as a member of the first particular set. The first particular storage device can be gracefully removed from the network, while only affecting the performance of the data access in the lower priority second particular set of storage devices.
According to another aspect of the invention, embodiments are provided in which the method includes determining whether a spare device is available for use in the first particular set of storage devices, and if a spare device is no
Parks Ronald L.
Taylor Alastair
Taylor James A.
Beausoliel Robert
Dell Products L.P.
Duncan Marc M
Hamilton & Terrile LLP
Terrile Stephen A.
LandOfFree
Method and apparatus for storage unit replacement according... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for storage unit replacement according..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for storage unit replacement according... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3074299