The storage management team is proud to announce the release of HP Storage Operations Manager (SOM) version 10.10, after SOM version 10.0 and 10.01.
So, here’s what we did in SOM 10.10:
Strengthen our offerings in analytics by introducing Storage Path Analytics and enhancing Virtual Server analytics
Support device events using SNMP traps.
Additions to device support matrix
Reporting solution moved to HP Operations Bridge 10.
And last but not the least, about the 30+ enhancements we put in based on customer feedback – yes, we listen!
Storage Path Analytics
One of the main value features of SOM 10.10 is the Storage Path Analytics for all hosts - including Inferred Hosts. This allows us to deliver significant value to the customer without having to communicating with the hosts. This allows you to manage large environments with relative ease and in very less time. There are 3 main use cases:
Identify Missing LUNS and/or storage paths to a LUN due to zoning misconfiguration: The path analytics on SOM analyze the storage volume presentations on the array and superimpose the fabric zoning configurations to highlight the set of LUNs and/or a set of paths missing to a LUN due to zoning misconfigurations. For example, a storage array may present volumes to a certain host, but if the host is not zoned with the array, then the storage will not be visible to the host.
Figure 1: Path analytics dashboard and drill-down for missing LUNs/LUN paths
Identify single points of failure in the storage paths: we identify Host Bus Adapter (HBA) ports and storage ports that are single points of failure in a storage path (meaning that if this port goes down, then the visibility to the LUN will be lost)
Figure 2: Path analytics dashboard and drill-down for single points of failure
Identify storage volumes shared by more than one host: Mostly, it is fine when volumes are shared by host cluster nodes. However, there is a risk of data corruption when a volume is shared by hosts that do not belong to the same host cluster. SOM dashboards can be used to identify hosts that are sharing volumes with other hosts and identify risks when they do not belong to the same cluster.
Figure 3: Path analytics dashboard and drill-down to identify shared storage volumes
Virtual Server Analytics
In SOM 10.01, we added 2 dashboards for Virtual Server Analytics, around presenting Thin Provisioning data – including datastore over-allocation and number of virtual servers with over-allocated datastores.
Figure 4: SOM 10.01 dashboard for virtual server analytics
With SOM 10.10, we have added more analytics around Virtual Servers:
1. Space Available for Growth for a datastore: This is the minimum of the space available in the datastore and the sum of the physical space available in the thin provisioning pools on the array(s) that are hosting the related storage volumes. The purpose of this view is to highlight that in a thin provisioned environment, a datastore may not necessarily have as much space to grow as seen from the virtual server.
Figure 5: Virtual server analytics showing space available for growth for a datastore
2. Virtual Machines by snapshot size and number of snapshots: The idea here is to track the space consumed by Virtual Machine (VM) snapshots in a given virtual server. We also track the number of snapshots created per VM. VMWare best practices (http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=1025279) advises that though there can be 32 snapshots in a chain, recommended is around 2-3. Managing this at an individual virtual server level could be easy, but when there are hundreds of servers, dashboards help simplify snapshot management.
Figure 6: Virtual server analytics showing VM snapshot data
The VM snapshot data is also shown in the VM Inventory view. This view can be sorted and/or filtered based on the above parameters.
Figure 7: VM inventory view showing snapshot details.
3. Capacity consumed by Powered off VMs: VMs consume disk space and hence it becomes important to track VMs that have been powered off for a while. By default, SOM dashboards highlight virtual servers by space consumed by VMs powered off for more than 3 days, across the datacenter.
Figure 8: Virtual Server Analytics showing space consumed by Powered off VMs
Support for device events using SNMP traps
As most of you know by now, SOM shares the same application server platform with Network Node Manager (NNMi). With SOM 10.10, we use the robust SNMP and Incident framework of NNMi to fetch traps from storage devices and switches. These traps are then converted into Incidents in the SOM. This release includes SNMP trap support for HP 3Par Storage Array, Brocade and Cisco Switches. We also support event correlation for switch port events to identify the device connected to the switch port using topology dependencies.
Additions to device support matrix:
SOM 10.10 comes with enhanced device content in terms of new device support for HPE StoreEasy, IBM SVC/V7000 and performance metrics for HPE XP7. We can also report on sub-lun tiering and deduplication for HPE 3Par devices.
HP Operations Bridge (OBR) 10 based Reporting Solution
SOM 10.10 reporting solution is based on Operations Bridge (OBR) version 10, which comes with a robust Vertica based platform. We have also simplified the OBR installation requirements by removing restrictions on Linux operating system group ID and mount points for installation location.
I’ll wrap up by inviting you to a webinar from Vivit, our independent user group. You can find more information on the webinar here.