storage Building Highly Available Systems: The Dell PowerEdge Cluster SE600W and PowerVault MD3000 By Nam Nguyen Daniel Moges Shabana M. Travis vigil G Related Categories: Clustering Dell PowerEdge servers The Dell™ PowerEdge™ Cluster SE600W is a Serial Attached SCSI (SAS)–based cluster solution comprising Dell PowerEdge servers, Microsoft® Windows® operating systems, and the Dell PowerVault™ MD3000 storage system. Deploying this solution enables enterprises to achieve highly available service levels in a cost-effective way. lobal commerce and the demand for 24/7 access Dell PowerEdge Cluster SE600W have made information availability vital to Dell PowerEdge clusters are based on Microsoft Cluster enterprise success. A lack of data access during Service software and designed to keep applications and maintenance, hardware or software failure, or any other services available during any single failure within the cluster. period of downtime can lead to decreased productivity, When Microsoft Cluster Service detects a failure, it automati- customer dissatisfaction, and lost revenue. As high- cally moves cluster resources from the failed cluster node to availability technology continues to advance, both the a healthy one and restarts the applications. PowerEdge clus- cost and complexity associated with clustering for avail- ters support cluster-aware applications such as Microsoft ability and failover have been reduced significantly, SQL Server™, Microsoft Exchange Server, and Oracle® Database making these configurations feasible for small businesses with Oracle Fail Safe software. as well as large enterprise-class data centers. In addition to application-level availability, enterprises For data to be continuously available, IT systems must should consider redundancy in the server-to-storage I/O maintain uninterrupted data access during both planned and path. Because a failure of any component along this path unplanned downtime. Host-based clustering, in conjunction (such as a server, adapter, controller, cable, or disk drive) High availability (HA) with high availability at the storage level, can greatly enhance jeopardizes system availability, the storage system’s ability Microsoft Windows Server 2003 IT service levels, and represents a crucial step toward ensur- to maintain data access during such a failure is a key part of ing data availability and minimizing downtime problems. a highly available storage deployment. Serial Attached SCSI (SAS) Allowing multiple servers (for example, a two-node cluster) The PowerEdge Cluster SE600W is the latest addition to shared access to data on a single storage system helps pro- the Dell family of feature-rich clustering solutions, and the vide application availability even if one of the servers first based on Serial Attached SCSI (SAS) technology. It cur- becomes unavailable. rently supports two-node clustering based on the compo- Dell PowerVault storage Direct attach storage (DAS) Storage architecture Visit www.dell.com/powersolutions for the complete category index. 70 The Dell PowerEdge Cluster SE600W solution and Dell nents shown in Figure 1. The supported PowerEdge servers PowerVault MD3000 storage array are designed to allow provide key high-availability features such as error-correcting enterprises of all sizes to build highly available systems in code memory; software or hardware RAID for the internal a cost-effective way. Deploying this solution can help main- drives; hot-swappable drives, power supplies, fans, and PCI tain access to critical application data even following a hard- slots; optional dual host bus adapters (HBAs); and optional ware or software failure. redundant paths to the storage systems. DELL POWER SOLUTIONS | May 2007 Reprinted from Dell Power Solutions, May 2007. Copyright © 2007 Dell Inc. All rights reserved. technology and offers high-availability features Operating systems Microsoft Windows Server 2003 and Windows Server 2003 Release 2 (R2) Enterprise Edition and Enterprise x64 Edition Servers Dell PowerEdge models 1800, 1850, 2800, 2850, 6800, 6850, 1950, 2900, 2950, 2970, and 6950 Storage Dell PowerVault MD3000 such as hot spares; hot-swappable drives, power supplies, cooling fans, and storage controllers; redundant storage controllers; redundant back-end paths with dual-port SAS drives; and optional premium features such as snapshots and virtual disk copy. To appreciate the potential value of cluster- Figure 1. Supported components for the Dell PowerEdge Cluster SE600W ing with PowerVault MD3000 storage, enterprises should first understand the host-based The PowerEdge Cluster SE600W offers three component fails, I/Os are transparently RAID method of clustering. This method uses a rerouted to the alternate path without affect- direct connection from a storage device to each ing applications or clients. server, and the RAID controller itself resides on Fully redundant configuration: This configu- a PCI RAID card installed in the server. In an offers only one path from each server to the ration provides the highest availability of external RAID configuration, like that of the storage (see Figure 2). If a component in one the three configurations, incorporating two PowerVault MD3000, the RAID controller resides path fails, the server loses access to the SAS 5/E HBAs and two paths from each outside the server in the external storage storage; when this happens, Microsoft server to the storage (see Figure 4). system itself. Cluster Service moves resources from the Administrators can configure I/Os to use Among the issues associated with host- affected server to the other server and any path for load balancing. If a storage based RAID clustering, one of the most signifi- restarts the applications. path fails, I/Os are transparently rerouted cant is related to the cache and performance. Redundant configuration with dual-port to the alternate path without affecting The cache is high-speed memory designed to HBAs: This configuration provides higher applications or clients. reduce the time required to read or write data. configurations: • • Entry-level configuration: This configuration • availability than the entry-level configura- In host-based RAID configurations, the cache tion, incorporating two ports on each SAS Dell PowerVault MD3000 resides on the RAID adapter card located in the 5/E HBA to create two paths from each server The Dell PowerVault MD3000 storage array, with server. To help ensure that in the event of a to the storage (see Figure 3). If an HBA fails, two high-performance, active/active external server failure the unwritten cache data is writ- Microsoft Cluster Service handles applica- RAID controllers and a mirrored cache, can be a ten to disk, vendors disable the use of the tion migration the same way it does in the key part of designing a cost-effective, highly cache on RAID adapters to maintain cache entry-level configuration. If another storage available cluster system. It is based on SAS coherency for the cluster. Unfortunately, doing Dell PowerEdge server LAN HBA Controller 1 Dell PowerEdge server Dell PowerEdge server HBA HBA Controller 2 Disk drives Dell PowerVault MD3000 Figure 2. Entry-level Dell PowerEdge Cluster SE600W configuration LAN Dell PowerEdge server HBA Controller 1 Controller 2 LAN Dell PowerEdge server HBA 1 HBA 2 Dell PowerEdge server HBA 1 Controller 1 Controller 2 Disk drives Disk drives Dell PowerVault MD3000 Dell PowerVault MD3000 Figure 3. Redundant Dell PowerEdge Cluster SE600W configuration with dual-port HBAs Reprinted from Dell Power Solutions, May 2007. Copyright © 2007 Dell Inc. All rights reserved. HBA 2 Figure 4. Fully redundant Dell PowerEdge Cluster SE600W configuration www.dell.com/powersolutions 71 storage so can cause a significant degradation in cluster performance. When using external RAID, the cache resides on the external controller within the storage system, and cache coherency can be maintained without disabling the cache functionality. For this reason, external systems are preferable for a clustered configuration. “The PowerEdge Cluster SE600W and PowerVault MD3000 are designed to build upon the reliability of parallel SCSI while addressing that technology’s performance, reliability, and scalability limitations.” Storing the cache separately from the server helps eliminate the risks associated with a server failure in a host-based RAID cluster. Cluster SE600W compared with the parallel As an external storage system, the PowerVault SCSI–based PowerEdge Cluster SE500W. MD3000 is ideal for highly available two- In a PowerEdge Cluster SE500W, the write cache is disabled in the RAID controller to maintain cache consistency, because the RAID intel- node clustering applications on PowerEdge Performance ligence is part of the controller residing on the servers when performance and access to data The PowerVault MD3000 can provide better host. In a PowerEdge Cluster SE600W, the write are critical. performance than PowerVault 22xS storage cache is enabled in the RAID controller, because because of its increased data transfer rate over the RAID intelligence is part of the controller Cluster features and advantages the SAS link and its enabled write cache. While residing on the external storage device. Each The PowerEdge Cluster SE600W and PowerVault the maximum data transfer rate for Ultra320 storage controller has 512 MB of cache, which MD3000 are designed to build upon the reli- SCSI is 320 Mbps (and the complications of the helps significantly improve overall system ability of parallel SCSI while addressing that shared bus architecture make increased speeds performance. technology’s performance, reliability, and scal- unlikely), SAS currently offers speeds of up to ability limitations. Figure 5 summarizes the 3 Gbps, and the SCSI Trade Association plans Reliability cluster features of the SAS-based PowerEdge for it to eventually support up to 12 Gbps. When deployed in an environment using parallel SCSI and a shared bus architecture, a misbehaving device can bring down the entire Dell PowerEdge Cluster SE500W Dell PowerEdge Cluster SE600W HBAs Up to two PowerEdge Expandable RAID Controller (PERC) 4/DC or PERC 4e/DC adapters per node Up to two SAS 5/E HBAs per node Storage Up to four PowerVault 22xS enclosures One PowerVault MD3000 enclosure connected to up to two PowerVault MD1000 enclosures Cluster SE600W configuration provides two paths Cables 68-pin shielded P-type SCSI cable for connection to the host, up to 12 meters long Compact cable and connector, up to 6 meters long dual-port SAS drives to allow data access from Hard drives 13 SCSI hard drives per enclosure 15 SAS hard drives per enclosure (with expansion to up to 45 drives by adding PowerVault MD1000 enclosures) As a result, if a component in a storage path fails, system. The PowerVault MD3000 offers a pointto-point architecture in which such devices can be easily isolated without affecting the entire system. In addition, the fully redundant PowerEdge from the servers to the storage, and includes either port from one of the two storage controllers. the multipath software can automatically reroute the I/Os to the alternate path without affecting Redundant path to the storage system No Yes, including redundant storage controllers with failover capability, dual-port SAS drives, and a multipath failover driver RAID RAID-1, RAID-5, RAID-10, and RAID-50 RAID-0, RAID-1, RAID-5, and RAID-10 Scalability Write cache Disabled in cluster mode Enabled in cluster mode the shared bus, while a SAS domain can sup- Storage manageability Dell OpenManage Storage Services and Dell OpenManage Array Manager, installed on a host directly connected to the storage Dell Modular Disk Storage Manager with advanced features such as snapshots and virtual disk copy, installed on a management station (either in-band through the SAS link or out-of-band over the network) port up to 16,384 devices without performance the applications or clients. Parallel SCSI can support up to 16 devices on degradation. The PowerEdge Cluster SE600W can support up to 45 SAS hard drives by adding PowerVault MD1000 disk expansion enclosures to the PowerVault MD3000 using a daisy-chain Figure 5. Feature comparison of Dell PowerEdge Cluster SE500W and PowerEdge Cluster SE600W 72 DELL POWER SOLUTIONS | May 2007 topology (see Figure 6). Reprinted from Dell Power Solutions, May 2007. Copyright © 2007 Dell Inc. All rights reserved. LAN Dell PowerEdge server HBA 1 HBA 2 Dell PowerEdge server HBA 1 HBA 2 Highly available cluster solution Daniel Moges is a systems engineer adviser in the The Dell PowerEdge Cluster SE600W and High-Availability Cluster Development Group at PowerVault MD3000 take advantage of SAS Dell. His responsibilities include developing SCSI-, technology to offer enhanced performance, reli- iSCSI-, and Fibre Channel–based high-availability ability, and scalability, and include additional clustering products; his current interests related to features not found in parallel SCSI–based high-availability clustering include enterprise stor- PowerEdge Cluster SE500W configurations. This age technologies and database systems. Daniel cost-effective clustering solution can benefit has a B.S. in Electrical Engineering from the enterprises of all sizes that require high avail- University of Illinois at Urbana-Champaign. ability for critical applications such as dataController 1 bases, messaging systems, and file, print, and Dell PowerVault MD3000 Shabana M. is an engineering analyst in the High- Web servers. Availability Cluster Development Group at Dell. Dell PowerVault MD1000 Nam Nguyen is a senior consultant in the High- and Fibre Channel–based clustering solutions and Availability Cluster Development Group at Dell, application performance. She has a B.E. in Dell PowerVault MD1000 and the lead engineer for SAS, Internet SCSI Computer Science and Engineering. Controller 2 Disk drives Her interests include the development of SCSI- Figure 6. Fully redundant Dell PowerEdge Cluster SE600W configuration with two daisychained Dell PowerVault MD1000 enclosures (iSCSI), and Fibre Channel Dell PowerEdge Cluster products. His current interests include business Travis Vigil is a senior product marketing consultant continuity, clustering, and storage technologies. for Dell PowerVault storage. He has a B.S. from He has a B.S. and an M.S. in Electrical Engineering Stanford University and an M.B.A. from Northwestern from the University of Texas at Austin. University’s Kellogg School of Management. Reprinted from Dell Power Solutions, May 2007. Copyright © 2007 Dell Inc. All rights reserved. www.dell.com/powersolutions 73