Unilog moves infrastructure to Google's public cloud

Company Overview

Unilog is a global technology company specializing in ecommerce and product data management in the B2B marketplace. Its flagship product is CIMM2, a software platform designed to facilitate commerce and communication between businesses. The company is headquartered in Bangalore, India with North American headquarters near Philadelphia, Pennsylvania. In 2015, Unilog was named one of the “50 Fastest Growing Tech Companies” by The Silicon Review.

Business Need

Unilog needed rapid deployments, unlimited scale, and reduced support costs for its customers. Unilog has multiple services (like managed infrastructure). Its clients are large ecommerce companies – meaning its small clients have about 500,000 SKUs on their website, and its large clients have about 8,000,000 SKUs. Unilog was struggling with administering and scaling client deployments to support its rapid growth (assistance in traditional infrastructure, cost, and support). Unilog approached Pythian because it needed experts to analyze its existing infrastructure, modify its design to support Google Cloud, validate its functionality, and provide platform, console, and escalation support.
Read MoreLess

Solution

Pythian’s first engagement with Unilog was to perform a lift and shift from its privately hosted environment to Google’s public cloud. We helped support an updated design of its legacy application stack, validated performance, then delivered production templates to accelerate its newly signed client deployments.   Unilog chose Google Cloud because of its cost, scalability, disaster recovery, security, and ease of use. Through three different Pythian engagements, Unilog is now leveraging GCS, GCE, GKE, and Cloud SQL.   Pythian moved Unilog’s Virtual Machines (VMs) and optimized the company for Google Cloud, specifically Google Compute Engine. Unilog has more than 100 customers on the platform, and 400-500 VMs. Pythian helped support its migration to Google in the following ways:
  • Analyze its existing physical & virtual systems
  • Architect its systems so they would run in an optimized way on GCP
  • Volume tested these configurations for validation and adjustments
  • Develop templates for each customer types (small, medium and large instances) that could be rapidly deployed for support
Pythian optimized Unilog’s existing systems to reduce operating costs, analyzing and directing it to stand-up a multi-tenant version. Through this modernized approach, it is able to support many customers across one application stack. This new approach allows them to stand-up clients more quickly, reduce support costs and consider a completely new SaaSbased licensing model that will greatly improve Unilog’s market value. Phase II of its migration to GCP extends the multi-tenant vision described above by extending the company’s architecture to leverage microservices which will reduce its total operating costs. This process consisted of the following:   GCP Optimization analysis consulting engagement considering utilization, spending and architecture. Our findings helped us develop an optimization plan which recommended the following broad-based actions:
  • Implement multi-tenancy for those customers who were the right profile
  • Re-architect its system to support microservices (PaaS) wherever possible
  • Replace Oracle with CloudSQL
  • Implement a containerized system (This is in process today)
  • Modernize its licensing and support models
Read MoreLess

Result

  • The initial migration saved Unilog up to 50% cost reduction for Infrastructure as a Service (IaaS) vs. AWS
  • Multitenancy improved their support costs by 30–40%
  • Multitenancy improved their operating margins
  • Replacement of Oracle will improve its gross profit
  • Embracing a micro services and containerization strategy is expected to reduce its operating costs by 60–70% versus a traditional IaaS model
  • Once its microservices strategy is fully implemented their operating costs will drop, system performance will improve, and features can be developed and implemented much more quickly
Read MoreLess

Explore Pythian’s popular services:

Technologies

  • Google Cloud
  • Google Cloud Storage (GCS)
  • Google Compute Engine (GCE)
  • Google Kubernetes Engine (GKE)
  • Cloud SQL

Looking to learn more about Cloud Migrations?