Can you share a list of problems that Hyper-Converged Infrastructure (HCI) helps resolve for an enterprise? If applicable, please provide some concrete examples.
Director Of Information Technology at Cass County Government
User
Top 20
2022-06-29T12:32:56Z
Jun 29, 2022
HCI's big sell is the single pane of glass management. No longer do you need to understand how to manage a separate storage environment and its firmware/software updates, worry about compatibility with new servers coming in, etc. It's all in one spot. I wouldn't go so far to say that you will need less IT staff to run the solution, but I would state you could get by with less 'specialized skillsets' within those staffers, as now it's all in the same place.
From a scaling side, it's easier to scale horizontally and has begun to be easier to scale vertically. If your storage/compute needs tend to be coupled together, an HCI option makes a lot of sense as when you add a node, you get both by default. If you, however, have more CPU-dependent workloads or need for high-density storage with low compute requirements, HCI tends to have some issues. However, every vendor seems to have recognized this, and now sell 'storage heavy' or 'compute heavy' nodes to help with these situations.
As for concrete examples, look no further than tools like Nimble DHCI. This is a fully integrated system, from networking to compute to storage, all supported by a single vendor. You no longer have to call Vmware, then storage, then network to troubleshoot an issue. All of your metrics and reports come from one spot. You can automate against a single API to configure and control the entire stack. If you need more resources, it's easy to add to a basic setup.
HCI is not for everyone. There are still plenty of places where a standard storage network, or a more highly advanced replicated storage/compute environment, or a cloud-focused/hybrid system make a lot more sense. It is, however, a great option for those that are looking for an easy-to-manage solution that is designed to scale with them when the time comes.
Hyper-Converged Infrastructure refers to a system where numerous integrated technologies can be managed within a single system, through one main channel. Typically software-centric, the architecture tightly integrates storage, networking, and virtual machines.
In my point of view, here are some points:
- One common system (no different vendors, dashboards, contracts, warranties, …).
- Centralized management reduces the need for IT expertise.
- Optimize server room capacities => reduce physical server hardware (reduce the cost of electricity, cooling and external IT resources).
- Start with a small environment: 1 or 2 nodes (with an HA cluster).
- Monitoring the hardware, running processes, CPU and memory usage, etc. are really easy now
- Always on deduplication and compression for all data
- Lower costs by providing more included services (no expensive upgrades).
- The main advantages that we feel are HA, easy resource allocation, Backup/Restore & of course the performance.
- A regional HUB can be the offsite disaster recovery for ROBO locations.
HCI Nodes can be interchanged between countries/regions in case upgrades/downgrades are needed.
Greets,
Hannes
HCI's big sell is the single pane of glass management. No longer do you need to understand how to manage a separate storage environment and its firmware/software updates, worry about compatibility with new servers coming in, etc. It's all in one spot. I wouldn't go so far to say that you will need less IT staff to run the solution, but I would state you could get by with less 'specialized skillsets' within those staffers, as now it's all in the same place.
From a scaling side, it's easier to scale horizontally and has begun to be easier to scale vertically. If your storage/compute needs tend to be coupled together, an HCI option makes a lot of sense as when you add a node, you get both by default. If you, however, have more CPU-dependent workloads or need for high-density storage with low compute requirements, HCI tends to have some issues. However, every vendor seems to have recognized this, and now sell 'storage heavy' or 'compute heavy' nodes to help with these situations.
As for concrete examples, look no further than tools like Nimble DHCI. This is a fully integrated system, from networking to compute to storage, all supported by a single vendor. You no longer have to call Vmware, then storage, then network to troubleshoot an issue. All of your metrics and reports come from one spot. You can automate against a single API to configure and control the entire stack. If you need more resources, it's easy to add to a basic setup.
HCI is not for everyone. There are still plenty of places where a standard storage network, or a more highly advanced replicated storage/compute environment, or a cloud-focused/hybrid system make a lot more sense. It is, however, a great option for those that are looking for an easy-to-manage solution that is designed to scale with them when the time comes.
You will get more automation, end-to-end visibility and you will need fewer people to run your IT solution.
It's also easier to scale...