Eliminate Configuration Vulnerabilities forEMC Symmetrix VMAX

EMC Symmetrix VMAX

You’ve surely invested a great deal of time and money in your EMC Symmetrix environment and carefully designed and implemented redundancy, availability and data protection. But what guarantees that your storage service will always deliver the required level of availability and efficiency that you expect?

It is difficult to maintain a configuration that is always aligned with vendor best-practices, and even harder to ensure that changes in other IT layers are well aligned with your Symmetrix storage arrays configuration. Unfortunately, even a minor gap between storage, compute and networking can put an entire business application at risk.

Symmetrix VMAX Configuration Vulnerabilities

A well-designed foundation is where it all starts or ends as the slightest misconfiguration can affect hundreds of servers and cause service or data loss issues. AvailabilityGuard performs deep storage array log analysis and checks for configuration issues or best-practice violations. Here are a few examples:

  • Front End adapter best-practices not met (inconsistent options, features, device mapping, improper load distribution between adapters and ports)
  • RAID group configuration best practices missed
  • Improper time synchronization
  • Hot spares are not configured correctly
  • Thin Pools and Pool Reserve Capacity (PRC) are not designed with sufficient reserves or grows abnormally
  • Meta-device configuration not aligned with best practices

FAST, FAST VP and FTS

Automating storage tiering carries tremendous benefits in terms of resource efficiency and consolidation, but must be carefully designed to insure the correct level of performance and resiliency. AvailabilityGuard proactively detects different types of issues that could affect both performance and resiliency. Here is a sample list:

  • Mixing rotational speeds per drive type in same pool
  • Performance Time Windows best practices are not met (e.g., avoid changing defaults when multiple applications are active on the array)
  • Misconfigured Fast Relocation Rate (FRR best practices)
  • Thin device binding not planned or implemented correctly
  • eDisk best practices in FTS missed
  • Configuration is not aligned with SRDF best practices for FAST

Symmetrix Configuration Alignment with Servers, Compute Clusters and Databases

Proper alignment of the storage layer to compute, cluster and databases is paramount. If key best practices are not met, serious data risks could arise, as well as application instabilities and significant performance degradation. These are a few examples of what AvailabilityGuard proactively detects:

  • Identify unauthorized access (e.g., non-cluster member accessing shared LUNs) and missing access (e.g., shared LUN not presented to all cluster nodes)
  • \Identify inconsistent or incorrect device mapping (e.g., inconsistent path configuration, port options, HBA-to-array compatibility, etc.)
  • Deep analysis of MPIO configuration (e.g., PowerPath, native MPIO, Symantec DMP and other) – including MPIO mode, load-balancing type, path status
  • Identify inappropriate storage tier allocation (e.g., using SATA with long stripes for swap devices and database log files)
  • Identify inconsistent or incomplete copies (Point-in-Time and replication – see more details below)
  • Identify cross-array configuration errors (e.g., LVM objects striped across multiple arrays; mirrored LVs kept on a single array, inappropriate consistency configuration and more)

Point-in-Time (PiT) Copy Validation

Point-in-Time copies allow extreme flexibility and play a key role in meeting important business needs such as quick restore from failure, I/O offloading and more. If configured improperly, your system could experience significant issues with data availability and consistency that would keep it from meeting those business needs. AvailabilityGuard analyzes Clones, Snapshots and BCVs (thin and thick) and validates:

  • Consistent split / ECA best practices (in particular, with conjunction to LVM and Database objects)
  • That RPO and retention goals are met
  • Identify stale and redundant replicas

Symmetrix VMAX Replication Considerations

AvailabilityGuard provides deep analysis of cross-array replication and mirroring, including SRDF (SRDF/A, SRDF/S, SRDF/Star), RecoverPoint and VPLEX – see more info on appropriate solution pages.