Rehost
Lift-and-shift onto EC2 (or equivalent) with as little change as possible. Fastest path out of a data centre when the app can tolerate it.
- AWS MGN-style replication
- Days to weeks per wave
- Rollback by failing back
Cloud Migration in Bangladesh
TiCON Cloud (TICON System Limited) migrates on-premise, VMware and other-cloud workloads onto AWS, and can also land them on Google Cloud or Azure when that is the right seller. Discovery comes first: inventory, dependencies and a TCO model, then a wave plan with a tested rollback. After cutover, the AWS estate is invoiced in taka through the reseller relationship — the same commercial path described on the pay-AWS page.
Last updated 27 September 2026
- 0–0 wk
- Typical small-to-mid migration
- 0 R's
- Strategy per workload
- 0–0%
- Typical compute & storage saving
- BDT
- Local invoicing after cutover
The 6 R's of cloud migration
AWS's large-migration guidance groups application decisions into strategies often called the 6 R's. Each workload gets its own R. Forcing every system through lift-and-shift, or every system through a rewrite, is how programmes stall.
Replatform
Lift and reshape: managed databases, containers, or a platform service without rewriting the business logic.
- RDS / Aurora
- EKS / ECS
- Lower run-cost than a raw VM
Repurchase
Drop the current product and buy a SaaS or equivalent that already runs the function.
- Email, ITSM, CRM class
- Data-export plan first
- Licence and residency check
Refactor
Re-architect for cloud-native scale — only where the business case pays for the engineering.
- Serverless / Lambda
- Event-driven pieces
- Longest of the six
Retire
Switch the workload off. Discovery often finds apps nobody logs into that still consume power and licences.
- Confirm owners
- Archive what audit needs
- Stop paying for it
Retain
Leave it where it is, on purpose — latency, a regulator, a vendor lock, or a rewrite that is not funded this year.
- Document the reason
- Review on a date
- Do not pretend it moved
Discovery, TCO and wave planning
Nothing production moves until discovery has a written output. The output is an inventory, a dependency map, a TCO comparison, and a wave list — not a slide that says “we will migrate.”
Discovery counts servers, licences, databases, file shares, identity, batch jobs and the humans who still know how a night job runs. Dependencies decide wave order: you do not cut over a payments app before the directory it authenticates against. TCO is the honest comparison of staying versus moving, including the partner invoice in taka, not only AWS list in dollars.
Waves are small enough to reverse. Each wave has a replication or copy method, a validation checklist, a cutover window, and a rollback. AWS Application Migration Service (MGN) is the usual rehost tool; AWS Database Migration Service (DMS) is the usual database path. Other R's use other tools. The method is chosen per wave, not once for the programme.
Regulated workloads — banks, government, telecom, health — need a residency and control conversation before the first replica is made. Bangladesh does not have an AWS Region. Data for those estates usually sits in ap-south-1 or ap-southeast-1, with the control set documented for the auditor. Circular numbers are not restated on this page; the data-residency article is the place for that once primary sources are attached.
How a TiCON migration runs
About 56 days
A reversible sequence. The quote comes after discovery, not before.
- Step 1:
Discovery
Inventory, dependency mapping and a TCO model. You get a written roadmap and a fixed quote for the agreed waves before anything production moves.
- Step 2:
Landing zone
A multi-account AWS foundation aligned to the Well-Architected Framework: identity, logging, network, backup and the billing relationship in taka.
- Step 3:
Wave execution
Prioritised, reversible cutovers. Replication where rehost is the R; platform or refactor work where that is the R.
- Step 4:
Validation
Functional, performance, security and cost checks against the checklist agreed in discovery. Rollback stays available until the wave is signed off.
- Step 5:
Stabilise and FinOps
Rightsizing, backup drills and a handover into managed operations if you want them. Savings Plans are a later decision, not a day-one requirement.
Downtime approach
This page does not promise zero downtime as a statistic. It promises a cutover plan that is designed for little or no user-visible downtime, with a tested rollback for each wave.
Replication-based rehost (for example AWS MGN) copies the workload while it is still running, then cuts over in a planned window. That is how most production servers move. Databases use continuous replication where the engine allows it. Some systems still need a freeze: a tightly coupled night batch, a vendor appliance, a licence server that cannot run in two places.
The honest sentence in the statement of work is which waves are near-zero and which waves have a declared window. “Zero” as a marketing stat on a whole estate is how programmes surprise the business at 2 a.m.
Typical timeline by workload size
These ranges are planning envelopes from how long discovery-plus-waves usually take, not a quote and not a sample bill. A regulated programme with a rewrite sits outside them.
| Size | What it usually means | Typical elapsed time after kickoff | Usual mix of R's |
|---|---|---|---|
| Small | A handful of servers or a single app, non-regulated | About 4–8 weeks including discovery | Mostly rehost, some replatform |
| Medium | Several applications, shared identity, one or two databases | About 8–16 weeks | Rehost plus replatform; a retire or two |
| Large | Multi-app estate, often regulated | About four to six months; some run longer | All six R's appear; refactor is scoped, not assumed |
Elapsed time is calendar time for a funded team, not person-days. Quotes follow discovery. Anonymised taka bills are not published on this page.
MAP and what the move costs in taka
Qualifying migrations can be assessed for the AWS Migration Acceleration Program. MAP is AWS's programme; TiCON handles the partner-side paperwork. Eligibility and any funding are set by AWS for that engagement. This page does not quote a co-funding percentage or a credit amount.
The professional-services price is a fixed quote after discovery. The run-cost after cutover is AWS usage on a taka invoice, typically 15–35% below list on compute and storage, and up to 40% lower on CloudFront bandwidth — those ranges are from TiCON's facts file. Use the BDT calculator for an indicative monthly run-rate; it is not the invoice.
Hidden costs that belong in the TCO, even without a sample bill: dual-running during waves, licence mobility, extra storage for replicas, and the people who still have to test. Omitting those is how a “cheap lift” becomes an overrun.
Frequently asked questions
How long does a cloud migration take in Bangladesh?
A small lift-and-shift is often 4–8 weeks including discovery. A multi-application mix is often 8–16 weeks. A regulated or refactor-heavy estate can take four to six months. Those are planning ranges, not a quote.
How much does cloud migration cost in Bangladesh?
It depends on workload count and which of the 6 R's each app needs. TiCON quotes in taka after discovery. Sample bills are not published here. MAP eligibility is assessed; funding amounts are not guessed.
Can you migrate with zero downtime?
We plan near-zero cutover with replication (for example AWS MGN) and a rollback per wave. Some systems still need a declared freeze. This page does not sell “zero downtime” as a whole-estate statistic.
Do you migrate from on-premise and VMware?
Yes. On-premise data centres, VMware, other clouds, and shared hosting are in scope. Databases typically move with AWS DMS when the engine allows continuous replication.
Can I pay for the migrated AWS workloads in BDT?
Yes. After cutover the AWS estate is invoiced in taka through TiCON's reseller relationship. You do not need an AWS dollar card to settle that invoice.
What are the 6 R's?
Rehost, replatform, repurchase, refactor, retire and retain. Each application gets one. AWS also documents a seventh, relocate, for hypervisor-level moves; we use it when it fits, without forcing it onto every wave.
Is AWS MAP funding available in Bangladesh?
Qualifying workloads can be assessed for MAP. AWS sets eligibility and any funding per engagement. This page does not publish a percentage of cost covered.
Sources
- About the migration strategies (6 R's and 7 Rs) (opens in a new tab). AWS Prescriptive Guidance. Accessed
- AWS Application Migration Service (MGN) (opens in a new tab). AWS Documentation. Accessed
- AWS Migration Acceleration Program (MAP) (opens in a new tab). Amazon Web Services. Accessed
- TiCON Cloud — AWS Partner Solutions Finder listing (opens in a new tab). Amazon Web Services. Accessed