HCI Isn’t the Only Simplified Management

Having decided that HCI is all about simplified deployment and management, I have started thinking about how simple it is to manage modern on-premises infrastructure. I feel that HCI is often compared with the technology it is replacing, rather than the alternatives that are available now. One aspect of that it when an HCI vendor says that a client replaced five racks of five-year-old equipment with half a rack of HCI. A new SAN and servers would not require five racks, so some of the reduced footprint is about hardware generations, rather than HCI. Another aspect is simplified management. HCI uses management that is centered around VMs and ideally policy based. Older architectures tend to have isolated control of each hardware type: compute, network, and storage. Comparing HCI’s simple management to a ten-year-old management practice is also not valid, HCI needs to be compared to the manageability of modern products with which HCI products compete.

Simplified Server Management

Simplified management often relies on APIs as a way to integrate the different parts of a data center stack under a common management tool. One of the things that Dell showed us at Tech Field Day 16 (disclaimer) was RedFish management of their servers. Redfish is an open standard API that is controlled by the DMTF (Distributed Management Task Force) as a way to manage a Software Defined Data Centre (SDDC). The Redfish API uses REST and JSON as its interfaces and allows much richer management than the IPMI standard that it supersedes. Dell talked about Redfish in their servers, implemented in the iDRAC alongside IPMI. With servers that support Redfish, it will be easier to build a software-driven data center, particularly one that has servers and other equipment from multiple vendors.

Simplified Hypervisor Management

In an HCI world, all management usually revolves around the virtualization management platform, with all of the other components slotting into the virtualization management console. The result is visibility gaps for the other teams, as they are often denied access by the virtualization team. Sometimes, HCI is deployed as a way to disenfranchise the storage team, removing them from storage management for the virtualization platform. A more constructive approach might be to have hypervisor management that integrates better with the tools used by other teams. I have not seen a lot of openness in hypervisor management. VMware has a published API for vSphere. However, they use a private API for their own management. It does seem that hypervisor management is one of the significant challenges in a multi-vendor SDDC.

Simplified Storage Management

Dell also talked about Swordfish, which is an extension to the Redfish API to manage storage systems. Swordfish is the successor to SMI-S (Storage Management Initiative – Storage). SMI-S has mostly failed to be the unifying management interface that it was hoped, vendor support has not been strong, and there is practically no multi-vendor management. The hope is that Swordfish will work as part of the wider use of Redfish.

VMware does have a strong play in simpler storage management for VMs; it is their VVols functionality. This allows storage management without VMFS and LUNs as was usual. VVols has VM-centric policy-based storage management. Precisely what we want for simplified management. Unfortunately, the uptake of VVols has been slow for both storage vendors and customers as well as being limited to vSphere based SDDCs.

Simplified Network Management

The network simplification that HCI delivers is the removal of a Fibre Channel storage network, HCI uses Ethernet for the storage network. In 2012 this was revolutionary, but now we don’t see so many new Fibre Channel deployments in small to medium data centers. The network continues to be the part of the data center where HCI does not offer integrated management; there is no physical switch management in any HCI product. Part of the problem is the lack of standardized methods for software configuration of physical switches. There is a network configuration section in the RedFish standard, so maybe we will see development in that direction. I would love to see the day when the network team use software to assign a network to a hypervisor cluster, the right switch ports automatically have the VLAN assigned, and all the hypervisor hosts automatically have a portgroup created with the correct VLAN and name.

One Manager to Manage them all

OK, not one manager for everything. We know that a single pane of glass is just vendor-speak for “we manage all of our parts in one tool.” The enterprise reality is multiple management tools from numerous vendors, what enterprises need is coordination between their various panes of glass. For example, the network team will want a network focussed management tool, but they need to be aware of what the virtualization team is doing. The network team definitely care which VMs reside on which VLAN. Using the same APIs as data sources for multiple consoles will make the data more consistent across the platforms. For example, the virtualization team may ask for nic3 & nic4 on host six to be connected to the HR network; the network team can use API integrated tools to see that those NICs plug into specific switch ports and then assign the right VLAN. The destination of this development is a system where the APIs are accessed directly from one management tool to another. Creating the port group on the virtualization platform will drive the API that assigns the VLANs to the right switch ports. To draw these all together, Dell wants to use Redfish as the core language of their OpenManage console, that is likely to take some time to develop.

More Integration Required

HCI does not deliver all of the simplified management that we would like and expect from an SDDC. It may well be that pre-HCI architectures make as much sense as HCI does, systems management has been improved over the five years since HCI started to be adopted. Don’t assume that an HCI product will be five times better than a multi-tier data center.

© 2018, Alastair. All rights reserved.

About Alastair

I am a professional geek, working in IT Infrastructure. Mostly I help to communicate and educate around the use of current technology and the direction of future technologies.
This entry was posted in General. Bookmark the permalink.