Phase 3: Data Migration Phase

In this phase, enterprise architects should ask followingquestions: What are the different storage options avail-able in the cloud today? What are the different RDBMS(commercial and open source) options available in thecloud today? What is my data segmentation strategy? What trade-offs do I have to make? How much effort (interms new development, one-off scripts) is required tomigrate all my data to the cloud?

When choosing the appropriate storage option, onesize does not fit all. There are several dimensions that youneed to consider so that your application can scale toyour needs appropriately with minimal effort. You haveto make the right trade-offs among various dimensions- cost, durability, query-ability, availability, latency, per-formance (response time), relational (SQL joins), size ofobject stored (large, small), accessibility, read heavy vs.write heavy, update frequency, cache-ability, consis-tency (strict, eventual) and transience (short-lived).Weigh your trade-offs carefully, and decide which onesare right for your application. The beauty about AWS isthat it does not restrict you to use one service or another.

Phase 4: Application Migration Phase

This phase determines if you should move part of or anentire IT infrastructure to the cloud without disruptingor interrupting your current business. There are twomain application migration strategies: Forklift or Hybrid.

Forklift Migration Strategy: Most suitable for state-less applications, tightly coupled applications, or self-con-tained applications. Rather than moving pieces of thesystem over time, forklift or "pick it all up at once" and move it to the cloud. In this approach, you might be able tomigrate an existing application into the cloud with minimal code changes. Most of the changes will involve copying your application binaries, creating and configuring Virtual Machine Images, setting up security groups andelastic IP addresses, DNS, switching to relational databasesin the cloud.

Hybrid Migration Strategy: Rather than moving theentire application at once, parts can be moved and opti-mized one at a time. This reduces the risk of unexpected behavior after migration and is ideal for large systems thatinvolve several applications. This strategy can be used to integrate cloud applications with other cloud-in compatible legacy applications (Mainframe applications or applications that require specialized hardware to function). In this strategy, you might have to design, architect and build tem-porary "wrappers" to enable communication between parts residing in your traditional datacenter and those thatwill reside in the cloud.