Migration to Hyper-Converged Infrastructure

Challenge: The demand on company services is doubling. PoPs are being added to improve network latencies. However, the IT infrastructure ability to support business growth lacks the agility to quickly respond. Multi-vendor, fragmented infrastructure resources are more complex to maintain, manage by a small IT staff. It is costly to operate, in the long run. Migrating the IT from Converge to Hyper-Converged Infrastructure (HCI) is investigated for potential datacenter migration.

Solution: As a Network Systems Architect, our SME managed multiple HCI vendors and helped the employing company identify a superior technical solution with the best five years Total Cost of Ownership (TCO). He researched the software and hardware technologies, potential vendors solution, and industry direction in data centers evolution. He developed HCI Feature Requirement Document (FRD) to serve as the base for RFP.

The sought solution shall provide infrastructure management simplification through a single-pane-of-glass to flexibly allocate compute, memory, network and storage resources on demand. It shall not be tightly coupled to a vendor hardware, rather the platform software and virtualization shall abstract the HW layer from the user.

Working with the employer team, the capacity and performance budgets showed that the demand on storage resources, from analytics engines such as Splunk, are likely to grow at a faster rate than the memory and compute resources. The addition of new storage capacity to the base solution shall not require the addition of servers rather storage units, in order to save CAPEX.

The system shall provide Recovery Point and Time objectives to deliver business continuity with minimal acceptable downtime. To balance performance and cost, the system shall have an optimized mix between SSD and SATA storage, as well as tested software intelligence to move data efficiently between hot and cold storage volumes.

With datacenter footprint is at a premium, the systems space utilization, per storage and compute resources, shall be high in order to accommodate future capacity growth within the existing COLO rack space during the project lifetime.

The platform software shall provide pervasive analytics to monitor resource utilization from the overall cluster-level, individual server unit, virtual layer, guests, and down to the application. The performance and capacity insights shall help forecast the need for adding new servers. The solution shall have open APIs to allow integration with north, south, east, and west interfaces for future automation, and enterprise level OSS/BSS.

Our SME developed the criteria for Bid Evaluation, and with stakeholders from the IT team and company operation, they established a scorecard composed of compliance to RFP FRD, TCO, SWOT analysis, PoC tests, as well as team assessment. He had a leading role in sales terms discussions, then created the hardware implementation document and project transition plan to IT team.