Rightsizing
Match instance, storage and database size to last month's actual use. Idle and orphaned resources are removed or stopped, not “optimised” in a slide.
- EC2 / RDS / EKS
- Idle cleanup
- Graviton where the workload allows
TiCON Cloud (TICON System Limited) runs and optimises AWS estates for teams that should be building product, not chasing tickets. Managed operations cover monitoring, patching, backups and incident response. FinOps is the practice that keeps the bill from drifting: rightsizing, commitments, and a monthly review with a named Technical Account Manager. Usage and the managed service arrive on one taka invoice.
Last updated 21 September 2026
The table is the default scope. Anything not listed is out of scope until it is written into the statement of work. AWS Enterprise Support is a different product; you do not have to buy it to have a TAM from TiCON.
| Area | Included | You still own |
|---|---|---|
| Monitoring & alerting | CloudWatch (and Grafana where agreed), on-call for infrastructure alerts | Application-level SLOs and product incidents that are not infrastructure |
| Patching | OS and selected data-plane patches on a written cadence, with a change window | Application releases and third-party SaaS you subscribe to directly |
| Backups & restore tests | Backup jobs, retention as documented, restore drills on an agreed calendar | Saying which restore point is the business “source of truth” |
| Security operations | IAM hygiene, secrets rotation where we hold the runbook, vulnerability scanning of in-scope hosts | Your identity provider, HR joiner/leaver process, and app-layer pentests you commission |
| Incident response | 24×7 response for in-scope infrastructure, with the TAM on the escalation path | Customer-of-your-customer communications unless we are contracted to speak |
| FinOps | Monthly cost review, anomaly follow-up, rightsizing and commitment recommendations | Approving a Savings Plan or Reserved Instance that commits your cash |
Numeric response times by severity are written in the managed-services contract. They are not restated as minutes on this page.
A public page that invents “15-minute P1” figures is a liability. Severity definitions belong in the contract, with the clock that starts when the alert fires and the humans who are on the rota.
What you can hold us to in writing: coverage is 24×7 for in-scope infrastructure, a named TAM in Bangladesh working hours with an escalation path outside them, and incident records you can show an auditor. What you cannot hold this page to: a minute count, a percentage uptime, or a penalty table. Those numbers, if they exist, are in the contract you sign.
Uptime of AWS regions and of a particular EC2 instance are different facts. AWS publishes its own SLAs per service. A managed-cloud SLA is about our response and our operations work, not a rewrite of AWS's SLA. If a region event is the cause, we work the AWS support path with you; we do not pretend we operate us-east-1.
A Technical Account Manager is a named person who knows the account. They are not the entire on-call rota, and they are not a substitute for your product owner.
The TAM joins architecture decisions, signs off the monthly FinOps review, and is the human you call when an incident has already been acknowledged by operations. They write the landing-zone exceptions, the backup-drill calendar, and the list of things that are explicitly out of scope. If the TAM is on leave, the runbooks and the rota still work — that is the test of whether the service is a person or a practice.
You meet the TAM in Dhaka or on a call. Offices also exist in Seoul and Tysons Corner; the invoice entity for Bangladeshi customers is the Dhaka company. Ask, in the contract, which legal entity bills you.
Most surprise bills are unused capacity, unattached storage, a forgotten environment, or traffic that never needed to leave the origin at full price. The monthly review is where those are named while they are still small.
Reseller pricing on compute and storage is typically 15–35% below AWS list, and CloudFront bandwidth is up to 40% lower. Those ranges are from TiCON's facts file. They are not a promise that FinOps will cut a given bill by 30–60%. A clean estate might save less; a neglected one might save more. The review shows the line items.
FinOps can be a standalone audit if you are not ready for 24×7 operations. The audit is time-boxed. Ongoing governance is a separate quote. Upload an invoice to the bill-analysis tool if you want a first pass before a call.
Match instance, storage and database size to last month's actual use. Idle and orphaned resources are removed or stopped, not “optimised” in a slide.
Savings Plans and Reserved Instances are modelled, then bought only when cash-flow and coverage justify them. They are not automatic.
Tags, budgets and anomaly alerts so a new environment cannot hide for a quarter.
We can help you prepare evidence for frameworks your auditors already named — SOC 2, ISO 27001, PCI-DSS — as a readiness engagement. That is not a claim that TiCON holds those certificates. The facts file has an empty ISO list; this page does not fill it from a guess.
What we actually operate day to day: least-privilege IAM, encryption in transit and at rest where the service supports it, secrets handling in a runbook, logging that an auditor can follow, and vulnerability scanning of in-scope hosts. What we do not do on a marketing page: invent a cert number, a surveillance-audit date, or a staff certification count.
DevOps engineering — Kubernetes, CI/CD, Terraform, VPC design — is available as project work on top of the managed retainer. It is scoped, quoted, and handed into the same operations team so the platform you built is the platform we watch.
EKS (or AKS/GKE when that is the estate) with Helm, autoscaling and observability handed into the same on-call.
Pipelines plus Terraform or CloudFormation so environments are reproducible, not click-ops.
VPC design, Route 53, ACM certificates, load balancers, WAF and CloudFront — the last of those at the reseller bandwidth rate.
Two lines on one taka invoice: AWS usage at reseller rates, and the managed-service fee. Both are quoted. There is no public retainer table on this page, because a three-server estate and a streaming platform are not the same job.
Reseller rates are typically 15–35% below AWS list on compute and storage, and up to 40% lower on CloudFront. Extra FinOps savings depend on waste in the current estate. This page does not publish a 30–60% promise.
24×7 monitoring of in-scope infrastructure, patching, backups and restore drills, security operations, incident response, a named TAM, and a monthly FinOps review. The scope table on this page is the default; the contract can add or remove lines.
Coverage is 24×7 for in-scope infrastructure. Response times by severity and any uptime percentage are in the managed-services contract. They are not invented as minutes on this page.
Yes. A time-boxed audit, then optional ongoing governance. You can start with the bill-analysis form if you only have an invoice to share.
Yes. Managed fees and AWS usage are combined on one taka invoice. You settle with TiCON, typically by bank transfer. Tax rates are NBR matters and are not restated here.
This page does not claim those certificates. We offer readiness support against frameworks your auditors name. A company cert list is unpublished until it is supplied as a sourced inventory.
Yes, as scoped project work that then hands into the same operations team. EKS, Terraform/CloudFormation, CI/CD and Lambda are in that catalogue.