AWS managed services provider in India for day-2 operations

AMS, an MSP programme partner or an independent provider?

Before comparing prices, find out which of the three you are being offered. The answer decides who answers at night, who signs the contract and what proof of process you can ask for.

No column wins by default. AMS suits enterprises that want AWS accountable end to end. A validated partner suits buyers who need the badge. An independent provider suits teams that care most about who does the work and want their AWS bill left alone.

When does validation by AWS actually matter?

Partner status changes what a provider can get from AWS. It does not change how carefully anyone patches your servers. It matters in four situations:

• Funding programmes. AWS’s Migration Acceleration Program (MAP) is delivered through AWS Professional Services and Migration Competency partners, and partner programmes carry their own incentives. An independent provider cannot apply for these on your behalf.
• Procurement rules. Some tenders and group IT policies name a partner tier. If yours does, that settles the question.
• One invoice for everything. If you want AWS usage billed by your provider, you need a reseller. We do not resell; your account stays with AWS India.
• Audited process. The MSP programme’s audit checks documented incident, change, security and billing processes. With an independent provider, ask to see those documents yourself.

To verify a claim, search the company in AWS’s Partner Solutions Finder. A genuine listing shows the programmes and designations AWS has validated. A logo on a website proves nothing; the directory entry does.

What day-2 operations on AWS should include

Day 2 is everything after launch. A scope worth signing names each area, how often it happens and what evidence you receive:

• Monitoring and on-call: alarms and synthetic checks tied to what users notice, a rota, and response targets by severity.
• Patching: baselines, maintenance windows, an emergency route for critical CVEs, and a compliance report listing the machines that missed a cycle.
• Backups and DR tests: plans per data store, copies outside the primary account or region, and restore drills with the measured time written down.
• IAM hygiene: single sign-on with MFA, no shared logins, unused keys removed, quarterly access reviews.
• Security findings: GuardDuty and Security Hub output triaged on a schedule, with a route to a managed SOC for real incidents.
• Cost: tags, budgets, anomaly alerts and a monthly review with named actions.
• Change management: changes made through infrastructure as code and reviewed pull requests, with a log of who changed what and why.
• SLAs: response and update times per severity, a written definition of P1, service hours, and the monthly report format.

Deeper reliability work such as SLOs, tracing and error budgets sits with SRE and observability. If the architecture has never been reviewed, an AWS Well-Architected review before handover gives you and the provider the same baseline.

Who does what: a responsibility table

AWS’s shared responsibility model separates security “of the cloud” (AWS: facilities, hardware, the infrastructure that runs its services) from security “in the cloud” (you: guest operating systems, security groups, identities, data). A managed services contract then splits your half again. This is the split we propose. R does the work, A is accountable, C is consulted, I is informed.

How AWS managed services are priced in India

Quotes from AWS managed services providers usually follow one of four models. There is no standard rate, so compare the written scope first and the price second.

We use the fixed monthly fee, set after a free consultation and a written proposal that lists scope, SLA and exclusions. Two costs sit outside any provider’s fee: your AWS usage and your AWS Support plan.

Do you still need an AWS Support plan?

Usually, yes. Only AWS can investigate a fault inside an AWS service or look at the underlying host. A provider opens and chases the case; AWS Support answers it. AWS replaced its plans on 2 December 2025. As of November 2026 the published pricing is:

You pay whichever is greater, the minimum or the percentage. Developer, Business (classic) and Enterprise On-Ramp end on 1 January 2027, according to AWS’s end-of-support notice, and Enterprise On-Ramp customers are being moved to Enterprise Support during 2026. If any account is still on a classic plan, budget for the change now.

What Indian buyers must check

Regions and where data sits

AWS has two regions in India: Asia Pacific (Mumbai), ap-south-1, opened in June 2016, and Asia Pacific (Hyderabad), ap-south-2, opened in November 2022 with three Availability Zones. Local Zones in Delhi and Kolkata are attached to the Mumbai region. A common design runs production in one region and keeps backups or DR in the other, so nothing leaves India. Newer regions do not carry every service, so check availability in Hyderabad before you design around it. Our Mumbai and Hyderabad pages cover the regional context.

Billing in rupees with GST

An account with an Indian contact and billing address contracts with Amazon Web Services India Private Limited, “AWS India” (AWS documentation; some AWS URLs still use the older abbreviation AISPL). AWS India bills in INR and charges GST. Sign-up includes identity verification that AWS says it carries out under CERT-In rules. Add your GSTIN to each account in Tax Settings so it appears on tax invoices. A provider should work inside this setup, not ask you to move accounts onto its own billing just to manage them.

CERT-In directions of 28 April 2022

The directions under section 70B(6) of the IT Act cover service providers, intermediaries, data centres, body corporates and government organisations. On AWS they translate into three jobs:

• Report the listed incident types to CERT-In “within 6 hours of noticing such incidents or being brought to notice about such incidents”. The provider’s runbook should produce the facts inside that window; your designated Point of Contact sends the report.
• Enable logs of all ICT systems and keep them “securely for a rolling period of 180 days” within Indian jurisdiction. In practice: CloudTrail, VPC flow, load balancer and application logs in an S3 bucket in Mumbai or Hyderabad, with lifecycle rules and Object Lock.
• Synchronise clocks with the NTP servers of NIC or NPL, or servers traceable to them. Entities spread across several countries may use another accurate source that does not deviate. Record which source your instances use.

DPDP Act and Rules

The DPDP Rules appeared in the Gazette on 13 November 2025. Consent-manager provisions begin on 13 November 2026; most duties, including security safeguards, one-year log retention under Rule 6 and breach notices to the Board, begin on 13 May 2027. A provider that can see personal data is your processor, so the contract needs processing terms. The DPDP rules checklist has the rule-by-rule detail.

RBI and SEBI regulated entities

RBI-regulated entities work under the Master Direction on Outsourcing of IT Services, issued on 10 April 2023 and in force from 1 October 2023. It covers cloud and SOC outsourcing, keeps the board and senior management responsible for outsourced work, and expects audit rights over the provider, a tested exit plan and Indian data storage where rules require it. SEBI-regulated entities follow the CSCRF circular of 20 August 2024. If either applies, settle audit and exit clauses before price. Some audits must be done by an empanelled auditor; Radiatus is not empanelled, and we say so before you sign.

Questions to ask an AWS MSP before you sign

1. Which of the three types are you? If you claim partner status, where is your Partner Solutions Finder listing?
2. Who is on call at 2 a.m. IST, by role, and how does an alarm reach them?
3. What are the response and update targets per severity, and what happens when you miss one?
4. Do your engineers sign in through named, MFA-protected roles we can revoke, or through shared keys?
5. Does every change go through code review and infrastructure as code? Where is the change log?
6. When did you last restore a client’s backup, and how long did it take?
7. How will you keep 180 days of logs in India, and who drafts our CERT-In report?
8. Who reviews cost, how often, and does your fee rise when our bill does?
9. Which AWS Support plan do you assume we hold, and who opens cases with AWS?
10. What do we receive when we leave: runbooks, diagrams, IaC, rotated credentials, and how many days of handover?

Red flags: a request for root credentials, a price quoted before anyone has looked at the accounts, an SLA with no definition of P1, “unlimited” support, and a partner badge you cannot find in AWS’s own directory.