It’s 2016 and still the number one storage management application is Microsoft Excel. Storage administrators continue to use Excel spreadsheets to track how much capacity each server is using and assign which LUN or volume goes to those servers. Of course the administrator has to enter all of this information manually when he has time. The problem is that the data center is too dynamic, by the “time” time becomes available the nature of that relationship is dramatically different.
Virtualization Changed Everything
Storage Management by Excel was almost reasonable before virtualization became the mainstay of the data center. The environment was more static and details like capacity and server to LUN relationships didn’t change very often. Even performance, which has always been variable, was manageable since there was dedicated storage resources given to each server.
Virtualization, and as data centers become more cloud like, breaks management by Excel. The environment becomes too dynamic. Applications can move from one server to another several times a day. Thanks to Storage vMotion data stores can be moved not only between LUNs, they can also move between physical storage systems. Not only is finding the time to update the Excel spreadsheet a problem, finding where volumes are may be impossible.
Performance management and monitoring also becomes more critical in virtualized environments since the storage system, the LUN and the storage network shares connections with other applications. Isolating a particular workload to track its performance over time and then logging that information in Excel is equally as impossible.
Silos of Storage Management
Storage vendors, of course, have their solutions and some of them are quite good, such as providing real-time analytics of their systems. The problem is these statistics are operating in a vacuum. They do not account for the impact of the storage network nor do they provide details up to the virtual machine.
The other problem with vendor-provided storage management tools is that they are siloed. They can only work with that vendor’s storage hardware and in some cases not even the vendor’s entire line of storage hardware. Most data centers, of course, have multiple storage systems for specific purposes they bought from a number of storage vendors. If they abandon Excel for storage management, they often have to resort back to it as a consolidator of the data that the vendor management tools put out.
Software Defined Storage Management
Software defined storage (SDS) is getting a lot of attention over the past few years. It attempts to solve the storage management problems by replacing the various vendors storage features sets with the SDS vendors feature set. The problem with this approach is customers often like the storage vendors feature sets, they trust them and have built entire processes around them.
The alternative to SDS is SDSM (Software Defined Storage Management), which instead of replacing the vendor’s storage services compliments them. SDSM applications provide a consolidation point for all storage performance data as well as reporting on various details like capacity consumption and LUN assignment. These solutions also provide an end-to-end view that encompasses the storage network switch, the storage host bus adaptor and even the virtual machine information.
The result is a single pane of glass you can use to not only replace the excel spreadsheet, but to enhance existing storage service via consolidated storage management. To learn more about banning Excel from storage management watch our latest ChalkTalk video with Storage Switzerland and SolarWinds.