Why Management is Critical for Data Center Lifecycle Management

Author: Dave BerryPMC Data Centers

A key tenant of cloud computing is that it’s an easy-to-deploy and manage infrastructure. Automating the management lowers costs and complexity when new infrastructure is deployed or changes need to take place.  Historically, the industry has deployed armies of administrators across the deployment lifecycle to make that happen, but that’s no longer practical or even necessary.

With the advent of cloud architectures, software-defined scale-out storage and virtualization layers, we need to automate all aspects of infrastructure management. Not just to realize the cost benefits of automation but to simplify overall management because there are more layers in a system.

As a result, all suppliers in the storage ecosystem—not just system suppliers—need to consider complete data center lifecycle management needs, including device-level, system/rack-level, and software-layer interfaces. Scoping the required management capabilities starts during product validation/qualification, factory burn-in, field support, operational discovery, configuration and monitoring, through to policy controlled IaaS orchestration.

This focus on increasing easy-to-use management capabilities was top of mind as we developed the 2015.1 software upgrade for PMC’s high-density 6G and 12G Adaptec® RAID adapters and HBAs. Here are a few of the recently released enhancements:

  • Upgraded ChipLink to support board-level products in addition to device bring up. ChipLink is a beloved tool for our customers during the validation and factory burn-in process to prove and enhance the quality of new infrastructure before deployment.
  • The PMC Baseboard Management Interface (PBSI)—A system-level health monitoring interface into a broad array of industry BMCs (Baseboard Management Controllers). The BMC a critical tool during the operational phase of managing infrastructure health and detecting errors.  PBSI presents an easy-to-use flat data set table that can be read and written, similar to any simple I2C device. It enables any BMC controller capable of supporting a I2C byte streaming interface.
  • High-Density Enclosure Support (HDES) – As we move towards rackscale architectures to manage the data deluge, data centers are connecting a lot more high-capacity JBODs. The HDES interface enables high density JBODs with multiple, cascaded SAS expanders, to support redundant cabling. It also includes expander module failover conditions for single and cascaded high density JBOD(s) and enhanced slot LED blink patterns to display more information about the RAID array status through enclosure LEDs. It also has a new mechanism to automatically store and clear enclosure slot and device target ID configurations, allowing dynamic enclosure and device configuration changes, as well as new scaling capabilities beyond existing support for SES 2/3, SGPIO (SFF-8489) and Virtual SES Enclosure (SFF-8488), and IBPI (SFF-8489).
  • Finally, orchestration—arguably what defines a cloud deployment because it provides the ability to configure and operate the infrastructure based on policies. VMware vSphere is, of course, an industry leader in this space. The PMC Adaptec Series 7 and Series 8 management stack now has plug-in support for this framework.

Just like running companies, departments and teams, management plays a critical role and can make or break execution. You can’t afford to overlook it.

This entry was posted by Carol Whitmarsh on at and is filed under Data Center. You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.

Leave a Reply

You must be logged in to post a comment.