Introduction
Multi-pathing software provides the intelligence necessary to manage multiple I/O paths between a server and a storage subsystem. Without multi-pathing software, the server operating system presents applications with multiple images of a disk or LUN (one for each I/O path discovered), which can result in data corruption.
At its most basic, multi-pathing software has two main modes of operation. When configured for redundancy, a single path is dedicated to I/O transfer, while other paths are in standby mode. The software manages failover between the I/O paths, thus eliminating the potential for a single point of failure. If connectivity along one path to a storage device is interrupted, the multi-pathing software dynamically switches I/Os to a surviving path, allowing application access to continue unimpeded. The other mode of operation allows for all paths to be utilized for I/O transfer. This can improve performance by leveraging the presence of these multiple paths, increasing the available bandwidth for I/O traffic.
Veritas Storage Foundation Dynamic Multi-pathing Option
Most multi-pathing solutions are developed by storage vendors for their specific brand of storage and, as such, generally cannot service the many various types of storage that exist in many of today’s heterogeneous SAN environments. Unlike these mostly proprietary solutions, Veritas Storage Foundation for Windows Dynamic Multi-pathing is a truly heterogeneous solution which fully integrates with the Microsoft® Multipath I/O (MPIO) architecture, including several Device Specific Modules (DSMs) which provide support for a wide variety of the most popular array families available today from the leading storage vendors. In addition to the benefits of heterogeneity, Veritas Dynamic Multi-pathing offers several advantages over other multi-pathing solutions, such as:
- GUI/CLI management for heterogeneous storage from a common user interface
- Array visualization from the GUI
- SNMP alerts for path failure and recovery
- Path performance statistics
- Microsoft WHQL logo qualification
- Fibre Channel and iSCSI support
Defining multi-pathing and load balancing
Dynamic Multi-pathing operates in two basic modes: Active/Passive mode, where a single path is dedicated to data transfer, while other paths act as failover targets to provide fault tolerance should the primary path fail; and Active/Active mode, where multiple paths are utilized for I/O transfer to provide improved performance through I/O load balancing.
Active/Passive
In its simplest form, Active/Passive multi-pathing is just that: one active path for I/O traffic, with other paths being passive. However, through special configuration, environments using Active/Passive multi-pathing can also be made to concurrently utilize multiple paths. This is known as Active/Passive Concurrent multi-pathing (also referred to as Dual-Active multi-pathing), and is important in cluster environments that use SCSI-2 protocols. This is accomplished through multi-pathing software that allows for configuring at the device (LUN) level instead of (or in addition to) at the array level. By configuring the preferred path for data transfer at the device level, specific paths can be dedicated to specific LUNs. This allows multiple paths, each configured as the preferred path for its specific LUN, to be used for data transfer. Note that the array also must support this.
Active/Active
Active/Active multi-pathing has more variations than Active/Passive. There are several different load-balancing algorithms available today. Although some of these algorithms may be referred to by different names, their functionality is the same. Table 1 describes some of these algorithms, along with some of their various other acronyms.
Written by: Jeffrey Armorer, Technical Product Manager, Storage and Availabiity Management Group
Download complete white paper below.
