What is our primary use case?
HPE Alletra Storage MP B10000 offers two types of storage solutions. One is MPX, which is object-based storage, and another one is MP B10000. B stands for block storage, and X is for object-based storage environments.
HPE Alletra Storage MP B10000 is the new 10000 series, whereas earlier versions were the 5000 and 6000 series, which are now out of sale. HPE Alletra Storage MP B10000 comes in two forms. One is 10000 B, which is pure SAN block storage, while another is MP 10000 X, the object-based storage. For AI purposes, X is the best fit if the data is related to that environment, while for virtualizations or databases, I recommend B, which is meant for structured data.
Regarding time-saving aspects, I can integrate HPE Alletra Storage MP B10000 with a total HPE environment. I can use the HPE VMware Essential option, which is side integrated and allows me to save money on VMware solutions. Earlier, since VMware was acquired by Dell, licenses became costlier based on previous licensing policies, but HP released a VMware solution to run different ERP applications. HPE Alletra Storage MP B10000 always provides better IOPS than others, allowing me to achieve the hypervisor layer on Alletra and storage and the servers, creating multiple VMs in that environment.
I have already deployed this in an IT company where I set up AI for classes by deploying HP servers and HP storage. On top of that, I installed the VMware hypervisor and then created the VMs, putting in multiple GPUs within the systems and connecting HPE Alletra Storage MP B10000 through the SAN switches only.
I deployed HPE Alletra Storage MP B10000 into a hybrid architecture with the help of VMware, allowing couple of VMs to move to the public cloud and come back from public cloud to on-premises.
In using HPE Alletra Storage MP B10000, the deployment depends totally on the workload. If the workload is hybrid for AI and for virtualizations or pure block-based architecture, it is easy to scale, which is one benefit of using HPE Alletra Storage MP B10000.
HPE Alletra Storage MP B10000 is meant for structured data, specifically for SAP, ERP, and Oracle databases, where high transactional structured data is needed. If the environment relates to PFS or AI, then I should use X instead of B.
I deployed an 18-node cluster, setting up two HPE Alletra Storage MP B10000 nodes over the SAN switches. Initially, for nine of the nodes, I ran VMware, and for eight nodes, I ran the Proxmox virtualization solution to create the cluster. It is easier and not complex, allowing me to migrate data running in the existing environment with the help of a third-party tool or, if the environment runs on VMware or Nutanix, I can migrate through the proprietary Move application of Nutanix.
What is most valuable?
Regarding HPE Alletra Storage MP B10000, its best feature is scalability. I can scale up to petabytes, and the data footprint size is lesser than others.
HPE Alletra Storage MP B10000 is easy to scale, available in both horizontal and vertical formats, and is always dependent on the controller level. I have to put it in the controller and then easily scale, allowing me to increase compute as well as disk simultaneously. In that environment, the IOPS will always increase in comparison to horizontally scaled.
From a maintenance and management perspective, HPE Alletra Storage MP B10000 has helped to reduce the amount of time I spend on storage management.
HPE Alletra Storage MP B10000 is basically a high throughput, high bandwidth solution that provides millions of IOPS. If the IOPS is high, then obviously the latency is very low, allowing me to run multiple VMs on that particular storage without any downtime. HPE Alletra Storage MP B10000 provides a guarantee of 100% availability.
HPE Alletra Storage MP B10000 provides 100% data availability. Earlier, this was only available with the Hitachi system, and now Alletra offers 100% data availability.
With the help of QoS functionality, I can assign limitations for dedicated IOPS-hungry applications. I can enable QoS to achieve the required IOPS for that particular application, and I do not think there is a challenge related to IOPS or performance.
The reason behind this is that HPE Alletra Storage MP B10000 is providing 100% data availability. If one controller fails, then data is accessible through the second controller.
What needs improvement?
I think there are some issues to enhance in HPE Alletra Storage MP B10000. In HP, there is no NAS as of now. Earlier, there was a NAS, but they stopped the unified boxes, creating challenges with customers. For enterprise customers, it is better to keep the environment separate, but for SMB customers, they should introduce unified boxes so that customers can easily use SAN/NAS functionality.
That function is not available in HPE Alletra Storage MP B10000. It is available in Dell PowerStore and NetApp. If the customer's requirement is pure NAS, in that environment, I cannot suggest HP solutions. If the customer asks for a unified box in an RFP, I have to go to alternative OEMs like NetApp.
As I mentioned, NetApp is known for the unified feature. In HP, there is no option to provide a unified box. If the requirement is for object storage or pure SAN, HP will best fit that environment. In NetApp, there are limitations on the SMB customer side due to the need for iSCSI switches between servers and storage. In contrast, with HPE Alletra Storage MP B10000, I can connect directly to the switches.
What do I think about the stability of the solution?
I have not received any performance issues with HPE Alletra Storage MP B10000. It is stable.
How are customer service and support?
I have never called HPE technical support, as most issues can be resolved through publicly available articles. For hardware issues, I raised a ticket, and since Alletra is a critical appliance, I have critical appliance support, which ensures that faulty hardware is replaced within a four-hour timeline.
In HP, I would rate the tech support a nine, as sometimes the queue length is lengthy, requiring me to hold the call for clarifications.
Overall, I would rate the tech support a nine.
Which solution did I use previously and why did I switch?
I have experience on other competitors like HP, so I worked on HPE Alletra Storage MP B10000, which is basically a competitor of StoreOnce appliances.
How was the initial setup?
HPE Alletra Storage MP B10000 deployment experience is not difficult. HPE Alletra Storage MP B10000 is much easier to deploy, and there is no challenge with the deployment of Alletra solutions.
What was our ROI?
ROI with HPE Alletra Storage MP B10000 depends on the use case, number of VMs, and infrastructure type. It cannot be justified solely on the number of nodes; it always depends on the use case.
Which other solutions did I evaluate?
Licensing in NetApp has two types: basic and advanced. The basic does not include functionalities such as snapshots, immutability, or ransomware protections, while advanced licensing supports data protection against ransomware.
What other advice do I have?
I do not have that option on this for managing and monitoring the global block storage environment from HPE Alletra Storage MP B10000's single cloud console.
I do not think I have an answer regarding whether HPE Alletra Storage MP B10000 enables line of business or DB admins to self-provision storage.
I do not have such data or examples on how much this ability has helped in lowering storage costs.
RPO/RTO should be minimal, and for that, I always have a DR solution to achieve that RPO/RTO.
If I require RPO/RTO zero with HPE Alletra Storage MP B10000, I have to synchronize, requiring two boxes with synchronized replication. Both storage units must confirm each other's status, and to mitigate RPO/RTO, I need to use synchronous and have both boxes available on dark fiber.
I would rate this review a 10.
Which deployment model are you using for this solution?
Hybrid Cloud
If public cloud, private cloud, or hybrid cloud, which cloud provider do you use?
Other