To Converge vs. Hyperconverge Infrastructure?

By John Rollason, Director of Product Marketing, Next Generation Data Center, NetApp

 

ESG White PaperEarlier today we shared that we are developing the next generation of hyperconverged infrastructure, built on SolidFire innovation. We will do what has not yet been done by the immature first generation of hyperconverged solutions — bring hyperconverged infrastructure to the enterprise by allowing customers the flexibility to run multiple workloads without compromising performance, scale, or efficiency. As part of the Data Fabric platform, it will also be the first fully cloud-integrated HCI solution, giving customers the ability to take advantage of their data across on-premises, public cloud, and hybrid cloud environments. With this innovation, we further broaden our industry-leading portfolio with cloud-integrated solutions available in standalone, converged, hyperconverged, and software-defined offerings.

 

About seven years ago when the concept of converged infrastructure (CI) began to emerge, CI proved to be simpler than purchasing servers, storage, and networking separately. This approach resulted in faster time to market and more successful data center deployment and management. Around the same time, adoption of VMware virtualization became the norm. Additionally, Amazon stopped just selling consumer goods and moved into web services. Why spend all that time and money on complex systems integration that adds no business value when others will do it for you? EMC got together with Cisco and VMware to launch the fully integrated vBlock. NetApp partnered with Cisco to launch the more flexible FlexPod. And everyone else copied. Converged infrastructure is now a $7b+ market.

 

The fundamental shift in how on-premises IT could be built was started by converged infrastructure solutions. Rapidly maturing cloud services have raised expectations further, with end users expecting to have access to new IT services within minutes, rather than days or weeks. This continues to impact the underlying infrastructure choices organizations need to make.

 

Hyperconverged infrastructure (HCI) appliances are in many ways a natural expansion of the success of the CI market, as customers continue to look for increased simplicity and convenience. HCI solutions promise to build on the ability of IT to more simply deliver predictable performance, speed up time to market, share hardware resources, and integrate solution vendor support. Physically, HCI solutions consist of software-defined appliances that combine server, storage, virtualization, and network connectivity resources into a single building block. HCI scales by adding new appliances that combine into a clustered, shared resource pool.

 

The first generation of HCI platforms have proved a great alternative for small/medium businesses (SMB) and remote office/branch office (ROBO) applications for organizations not fully ready to embrace a wholesale move to public cloud and SaaS services. These first generation HCI offerings have also gained early success for smaller scale virtual desktop infrastructure (VDI) projects. However, within the core data center, HCI solutions have not yet proved a match for more mature CI and 3-tier solutions due to limitations around performance, automation, the ability to share workloads, lack of scale, and limited configuration flexibility.

 

Today’s HCI solutions typically are purchased with a fixed ratio of compute-to-storage resources, which adds to application licensing costs and limits their utility for mixed-workload environments. While many are marketed as on-premises “cloud solutions,” they lack the ability to guarantee performance for shared or mixed workloads and become very expensive and inefficient as they scale. They often lack mature data management and data services, resulting in significant deployment limitations. They are not integrated into mature CI, hybrid cloud, or Data Fabric strategies, also restricting their use to isolated, smaller, on-premises workloads. This has mostly limited HCI to smaller-scale clusters of 4-8 nodes, resulting in many costly HCI “silos” as demands grow.

 

There is no doubt the market for HCI is exciting, diversifying, and growing extremely fast. Demand will only increase for the promise of a vastly simplified solution for on-premises IT infrastructure. Just as all vendors offer converged solutions today, they will all no doubt offer HCI. Most organizations will continue to require a mix of consumption models, from cloud services to software-only, with CI, HCI, and traditional appliances in between. As customers select technologies, perhaps the most important factor is to choose to partner with a vendor that will help guide them toward the Next Generation Data Center, whatever the service and deployment model requirements may be.

 

Gain further insight and learn from the latest ESG survey on converged vs. hyperconverged. Download the white paper, “Seeing Past the Hype: Understanding Converged and Hyperconverged Platforms.

Comments

Hello Experts,

 

I am in process to build Graphana, Graphite & NetApp Harvest tool in my storage environment.  Following are some information about my new setup.

 

     # Installed NetApp Harvest Tools on RHEL 7.x host.

     # Installed/configured Graphite & Graphana on another RHEL 7.x host.

 

While importing the default dashboard to Graphana from Harvest server, I am getting below error.  Not sure why ?

 

=============================

[OK ] Will import dashboards to [https://localhost:3000]
[OK ] Dashboard directory is [/opt/netapp-harvest/grafana]
[ERROR ] Unable to establish TCP connection to localhost:3000 [127.0.0.1:3000]
[ERROR ] Exiting due to fatal error.

==============================

 

I am using installation guide (NetApp Harvest Installation and Administration Guide 1.3) & installed below package on host.

 

netapp-harvest-1.3-1.noarch

 

Thank you in advance !!!

 

Regards,

Binod