Home›Cloud›Operations›CloudOps

XcellOps|CloudOps

Build it as code. Watch it against an SLO. Fix it under ITIL. Pay less for it every month.

CloudOps is the operating model that keeps AWS, Azure, Google Cloud, Oracle Cloud and XcellHost's Mumbai private cloud provisioned, observed, patched, backed up, right-sized and reported — run by our 24×7 NOC and SOC.

Provisioning & IaC
Monitoring & observability
Incident, problem & change
Patching & backup checks
FinOps in rupees
SLOs & error budgets
View packages → Savings estimator SLO maths
✦Ask AI about CloudOpsChatGPTClaudePerplexityGoogle AICopilot
24×7 NOC + SOCMumbai · ITIL 4 practices
Plan reviewedno apply without approval
CloudOps control plane · provision IST --:--
01IaC plan → apply Sample data
Resources0
Clouds0
Drift—
Powered by XcellHost CloudOps · Mumbai Run 24×7 by XcellHost NOC + SOC
ISO 20000-1 & 27001SOC 2 Type I & IIMeitY empanelledTier IV DC · Mumbai24×7 NOC + SOCSince 1999
Overview

CloudOps is how a cloud gets run, not just built.

Migration ends on a date. Operations never does. CloudOps is the set of practices, tools and people that keep every cloud estate healthy, secure and affordable after go-live.

Most Indian businesses now run a mix: a hyperscaler for new applications, Microsoft 365 and Azure for the workplace, and a private cloud for ERP or regulated data. Each brings its own console, alerts, patch cycle and invoice. Without one operating model, changes go in by hand, alerts are ignored, nobody owns the bill and audit evidence is gathered the night before.

XcellHost CloudOps replaces that with a single loop: infrastructure defined as code, observed against service level objectives, incidents, problems and changes handled under ITIL 4 practices, patching, backup and capacity on a calendar, and spend reviewed every month using the FinOps Inform → Optimize → Operate cycle. Our Mumbai NOC and SOC run it around the clock.

Looking for the managed-service contract detail — RACI, SLA targets and per-cloud scope? See Managed Cloud Operations. This page explains the CloudOps model and lets you size it.

One loopprovision → observe → respond → improve
Five cloudsAWS · Azure · Google Cloud · OCI · XcellCloud
Two clocksSLO for users, SLA for the contract
One invoiceINR, with GST, from an Indian company
CloudOps lifecycle

Eight stages. One loop that never stops.

Click any stage on the ring. Each turn of the loop feeds the next: an incident becomes a problem record, a problem becomes a change, a change becomes code.

01 Provision as code
01ProvisionChange practice

—

    Output you can audit—
    Capabilities

    What our engineers actually do in CloudOps.

    Ten capabilities, each mapped to an ITIL 4 practice or an SRE/FinOps discipline. Every one is in scope from day one; depth depends on the package.

    Provisioning & IaC

    Landing zones, networks and compute defined in Terraform or OpenTofu*, reviewed as pull requests and applied from a pipeline. Drift is detected and reported.

    Change enablement

    Monitoring & observability

    Metrics, logs and traces with OpenTelemetry where it fits. Alerts on symptoms users feel — latency, errors, traffic, saturation — not on every CPU spike.

    Monitoring & event mgmt

    Incident, problem & change

    ITIL 4 practices run inside XcellHost's ISO 20000-1 certified service management system: priorities, bridges, root-cause records and a change calendar.

    ISO 20000-1

    Patching

    OS and platform patches in agreed windows, tested on non-production first, with a compliance report. Deeper endpoint and third-party coverage via Patch Management as a Service.

    Infrastructure & platform mgmt

    Backup & restore tests

    Snapshot and backup policies checked daily, restores tested on a schedule. Failover to Mumbai via Cloud Disaster Recovery.

    Service continuity

    Capacity & performance

    Trends for compute, storage, database connections and quotas, forecast before month-end or a sale day. Autoscaling limits tuned to real load.

    Capacity & performance

    FinOps

    Tagging, showback, idle clean-up, rightsizing and commitment planning, following the FinOps Foundation's Inform, Optimize and Operate phases. Reported in rupees.

    FinOps Framework

    SRE · SLOs & error budgets

    Service level objectives agreed per service, error budgets tracked daily and burn-rate alerts that pause risky changes before users notice.

    Service level mgmt

    Multi-cloud operations

    One alert standard, one tagging policy, one change calendar and one report across AWS, Azure, Google Cloud, Oracle Cloud and XcellCloud.

    Service configuration mgmt

    Automation & runbooks

    Repeat fixes written as tested runbooks — restart, clear, scale, roll back — with approval gates. Correlation and auto-remediation at scale via AIOps.

    Shift-left automation
    Multi-cloud

    Five clouds. One operating standard.

    Native tools stay where they are best; the operating model on top is the same. Pick a cloud to see what we use and what we run.

    Amazon Web Services*

    Managed AWS →
    Provision

    —

    Observe

    —

    Optimise

    —

    Guard

    —

    —

    Tool · cloud cost-savings estimator

    Where could your cloud bill come down?

    Three levers, applied in order so nothing is counted twice: remove waste, schedule non-production, then commit on steady usage. Illustrative ranges — the formula is shown below the result.

    Main hyperscaler (sets the published commitment ceiling)
    Illustrative monthly saving
    ₹0to₹0

    0–0% of today's spend · about ₹0 a year at the midpoint

    Waste removed₹0
    Non-prod scheduled₹0
    Commitments₹0
    vs CloudOps Professional + FinOps₹39,999 / month*
    Formula, assumptions & sources
    1. Waste = spend × idle share × 60–90% (assumption: not every oversized resource can be cut at once).
    2. Scheduling = (spend − waste) × non-prod share × (168 − hours on) ÷ 168 × 70–100% (assumption: some non-prod must stay up).
    3. Commitments = remaining production spend × eligible share × (1 − covered share) × discount range. Range used: 20% to 72% × 0.6. Published ceilings: AWS Savings Plans up to 72% off On-Demand; Azure Reservations up to 72% on VMs and savings plan for compute up to 65%; Google Cloud resource-based CUDs up to 55% (70% memory-optimised). Real discounts depend on term, payment and instance family.

    Sources: AWS whitepaper "Cost optimization — reservation models" (Savings Plans); Microsoft Azure Reservations pricing page; Google Cloud "Committed use discounts overview" — all read November 2026. Not a quote; a billing-data review gives the real figure.

    Tool · SLO & error-budget calculator

    What does 99.9% really allow?

    Exact maths: allowed downtime = (1 − SLO) × the length of the window. Enter what has happened so far this window to see how fast the budget is burning.

    Availability SLO
    Window
    SLOper dayper weekper 30 daysper year
    Error budget for this window
    43 min 12 sof allowed downtime in 30 days
    0%budget used
    Failed requests allowed0
    Burn rate0×
    Budget left0
    At this pace, exhausted on—

    —

    Burn rate = share of budget used ÷ share of window elapsed. 1× means the budget lasts exactly the window. A common SRE practice is to alert on fast burns and pause risky changes when the budget is spent.

    CloudOps maturity model

    Five levels. Find yours, then the next step.

    A practical model we use in assessments. Pick the level that sounds most like you today.

    What it looks like

    —

    Typical signals
      Next step with XcellHost

      —

      Configure →
      How XcellHost delivers

      An operating rhythm, not a ticket queue.

      The same certified team runs every CloudOps package. What changes is the cadence you see.

      Every minute

      Watch

      • 24×7 monitoring from the Mumbai NOC
      • Security alerts triaged by the SOC
      • P1 bridge opened on detection
      Every day

      Check

      • Backup jobs and snapshots verified
      • Error budgets and burn rates read
      • Drift and idle resources listed
      Every week

      Change

      • Change review and calendar
      • Patch window on non-prod, then prod
      • Ops call (Professional and above)
      Every month

      Report

      • Service report: uptime, incidents, changes
      • FinOps review with savings in ₹
      • Patch and backup compliance
      Every quarter

      Improve

      • Business review and capacity outlook
      • Restore or DR drill
      • Automation backlog you approve
      24×7 NOC & SOCMumbai, staffed around the clock
      Tier IV data centreXcellCloud private cloud, Mumbai
      ISO 20000-1 · 27001also ISO 9001, 27017, 27018
      SOC 2 Type I & IIMeitY empanelled
      One INR invoiceGST from an Indian company
      India rules the loop is built around CERT-In directions (April 2022): report specified incidents within 6 hours and keep ICT logs for a rolling 180 days in India — central logging and incident runbooks support both. DPDP Act 2023 and Rules 2025: Indian regions by default, logged access and fast breach facts. Core duties apply from 13 May 2027. Orientation only, not legal advice; compliance stays with your organisation.
      Toolchain integrations

      We work with the tools you already pay for.

      Filter by job. Where you have nothing, we recommend the lightest option that fits — open source first where it does the job.

      Terraform*OpenTofu*Ansible*Pulumi*GitHub Actions* / GitLab CI*AWS CloudFormation*Azure Bicep* OpenTelemetryPrometheus*Grafana*Amazon CloudWatch*Azure Monitor*Google Cloud Observability*Leading APM platforms* ServiceNow*Jira Service Management*Zoho Desk*On-call paging tools* AWS Cost Explorer*Azure Cost Management*Google Cloud Billing*FOCUS billing formatTag policies Kubernetes*Red Hat OpenShift*VMware*OpenStack*Proxmox* Veeam*Acronis*Commvault*Native snapshots

      * Product names are trademarks of their owners and are used only to describe compatibility. No partnership or endorsement is implied unless stated on the vendor's own XcellHost page.

      Packages

      Priced per cloud estate. Same tiers as Managed Cloud Operations.

      One estate = one AWS organisation, Azure tenant, Google Cloud organisation, OCI tenancy or XcellCloud project. Cloud usage is billed separately at actuals.

      01 · Essential

      Run it safely

      ₹14,999/ estate / month*

      up to 10 resources · extra ₹999* each

      • 24×7 monitoring & alerting
      • Incident response, P1 24×7
      • Monthly patching window
      • Daily backup job checks
      • Monthly service report
      Most chosen 02 · Professional

      Run it as CloudOps

      ₹39,999/ estate / month*

      up to 40 resources · extra ₹999* each

      • Everything in Essential
      • L2 cloud specialists, P1–P2 24×7
      • Change & problem management
      • Guardrails & posture triage
      • FinOps review every month
      • Weekly ops call · QBR
      03 · Enterprise

      Run it at scale

      On quotescoped after assessment

      regulated or large multi-cloud estates

      • Everything in Professional
      • Named service manager
      • SLO & error-budget programme
      • Regulator-ready evidence pack
      • DR drill support

      * Indicative — confirm with XcellHost price list. INR, excl. 18% GST. Tier prices, extra resources (₹999*) and onboarding (₹24,999* one-time) are the same as on Managed Cloud Operations; add-on prices are taken from Managed Google Cloud, Managed AWS, Patch Management and AIOps pages. The SLO & error-budget programme (₹19,999*/month) is a new placeholder to confirm.

      Configure

      Build your CloudOps package — see the rupee total.

      Fixed-price lines go to the cart at price × quantity. Enterprise comes with a quotation.

      1 · Tier
      4 · Add-ons
      Your package · indicative

      Professional · 1 estate

        Monthly₹0
        One-time₹0
        Subtotal₹0
        GST (18%)₹0
        First invoice incl. GST₹0

        * Indicative — confirm with XcellHost. INR excl. 18% GST until the last lines. Cloud consumption billed separately at actuals.

        FAQ

        CloudOps, answered.

        CloudOps — cloud operations — is the practice of running cloud infrastructure day to day: provisioning it as code, monitoring and observing it, handling incidents, problems and changes, patching, backup, capacity, cost management and reporting. It is what keeps a cloud reliable and affordable after migration.

        Managed Cloud Operations is the service contract: scope per cloud, RACI and SLA targets. This CloudOps page explains the operating model behind it — IaC, SLOs, FinOps, maturity — and lets you size a package. The tier prices are the same.

        Amazon Web Services, Microsoft Azure, Google Cloud, Oracle Cloud Infrastructure and XcellCloud, XcellHost's private and public cloud in our Mumbai Tier IV data centre — plus VMware, OpenStack and Proxmox estates. Several can run on one desk and one report.

        An SLO is an internal reliability target for a service, such as 99.9% of requests succeeding over 30 days. An SLA is the contractual commitment, with remedies if it is missed. SLOs are usually stricter than SLAs, so the team reacts before the contract is at risk.

        Over a 30-day window, 99.9% allows 43.2 minutes; 99.95% allows 21.6 minutes; 99.99% allows 4.32 minutes (about 4 minutes 19 seconds). Over a 365-day year, 99.9% allows about 8 hours 46 minutes. The calculator above does the maths for any window.

        We follow the FinOps Foundation's three phases. Inform: tag, allocate and show spend by team. Optimize: remove idle resources, right-size and plan commitments. Operate: set budgets, alerts and policies so savings stick. Results are reported in rupees every month.

        Yes. Incident, problem, change enablement, monitoring and event management, and service level management follow ITIL 4 practices inside XcellHost's ISO 20000-1 certified IT service management system. Records are kept in your ITSM tool or ours.

        No. Normal changes are approved by you. Low-risk repeatable changes, such as a scale-out within agreed limits, can be pre-approved as standard changes. Infrastructure-as-code changes are reviewed as pull requests before apply, and every action is logged.

        Essential is ₹14,999* and Professional ₹39,999* per cloud estate per month, with ₹999* per extra resource and ₹24,999* one-time onboarding. Enterprise is on quote. All indicative, excl. 18% GST; cloud usage is billed at actuals.

        Central logging with 180-day retention in India, clocks synced to Indian NTP sources and an incident runbook support CERT-In's directions, including the 6-hour report. Indian regions by default, logged access and quick breach facts support DPDP duties. The legal obligations remain yours.

        Send us one month's cloud bill. We'll send back a CloudOps plan.

        A free 30-minute review: where the waste is, which SLOs matter, and a package price in rupees within two business days.*

        See packages → WhatsApp
        CLOUDOPS RUN
        BY XCELLHOST
        ISO 20000-1 · ISO 27001SOC 2 Type I & II24×7 NOC & SOCTier IV · MumbaiSince 1999