Grafana Cloud — Environment Compatibility & Sizing Brief

Prepared for prospect environment assessment  ·  Rene Wong, Strategic Channel Manager

PRE-SALES · CONFIDENTIAL

1   Stack Compatibility

How Grafana Cloud maps to each component in the prospect's environment. All integrations are configured via the Connections UI and use Grafana Alloy for collection.

6/10
Turnkey or JVM
Coverage
4/10
Custom / Infra-Layer
Monitoring
100+
Total Integrations
in Catalog

Turnkey Integrations (out-of-the-box dashboards, alerts & scrape configs)

ComponentIntegrationCoverageWhat You Get
Apache HTTP Server Apache HTTP Server ✅ Turnkey Request rates, response codes, worker status, connection metrics, OOB dashboards + alerts
Elasticsearch Elasticsearch ✅ Turnkey Cluster health, indexing rate, search latency, JVM heap, shard status
PostgreSQL PostgreSQL ✅ Turnkey Connections, replication lag, query stats, table bloat. Upsell: Database Observability for query-level insight
RabbitMQ RabbitMQ ✅ Turnkey Queue depth, publish/consume rates, connection counts, node health

JVM-Level Coverage

ComponentIntegrationCoverageWhat You Get
JBoss EAP v8 Java JVM ☕ JVM Heap, GC, thread pools, class loading. JMX metrics from EAP management subsystem via Alloy's JMX receiver for app-server-specific metrics (deployed apps, datasource pools, HTTP connector stats)
Newgen Services Java JVM (if Java-based) ☕ JVM Newgen typically runs on Java — JVM integration covers heap, GC, thread pools. Custom metrics via Alloy's Prometheus scrape if the app exposes a /metrics endpoint. Logs → Loki.

Infrastructure-Layer Monitoring (no turnkey integration — fully observable via Alloy + Loki)

ComponentCoverageHow to Monitor
Kibana 🔧 Custom Expose Kibana's /api/stats or Prometheus endpoint → scrape with Alloy into Mimir. Build dashboards on request rate, response time, concurrent connections. ELK consolidation opportunity: Grafana can replace Kibana for visualization.
MS-SQL DB 🔧 Custom Alloy scrapes sql_exporter or MSSQL Prometheus exporter. Custom dashboards for wait stats, query duration, buffer cache, deadlocks.
OnlyOffice 🔧 Custom Container/host metrics via Linux/Docker integration. HTTP endpoint checks via Synthetic Monitoring for uptime & latency. Application logs → Loki.
ImplicitSync Licensing 🔧 Custom Host-level metrics via Linux integration. Application logs → Loki. Custom metrics via Alloy Prometheus scrape if the app exposes a /metrics endpoint.

💡 ELK Consolidation Opportunity

This environment runs Elasticsearch + Kibana — likely for log search. Grafana Cloud with Loki offers log aggregation at a fraction of the cost (no indexing tax), with logs ↔ metrics ↔ traces correlation in one pane. Grafana replaces Kibana for visualization. This is a natural consolidation play.

2   Sizing Questions for Budgetary Estimate

Answers to these questions let us model a ballpark annual cost. Share with the customer's platform/infra team.

Metrics (Mimir)

1How many hosts / VMs / containers run these services?
Drives active-series estimate. Each Linux host ≈ 500 series; each container ≈ 200 series.
2What scrape interval do you use (or would target)?
Billing = active series × DPM. 60s = 1 DPM, 30s = 2 DPM, 15s = 4 DPM. Non-critical workloads at 60s cuts cost 4×.
3Are you already running Prometheus? If so, how many active series?
Direct input to pricing. Check Prometheus TSDB status page or prometheus_tsdb_head_series.
4How long do you need to retain metrics?
Free = 14 days, Pro = 13 months, Enterprise = custom. Longer retention is a policy choice, not a disk problem.

Logs (Loki)

5What is your monthly log ingest volume (GB/day or GB/month)?
Loki pricing: Process $0.05/GB + Write $0.40/GB + Retain $0.10/GB/mo. Shorthand ≈ $0.55/GB at 30-day retention.
6Which services generate the most log volume?
Identifies filtering / Adaptive Logs opportunities. JBoss + Elasticsearch tend to be noisy.
7How long do you need to retain logs?
Pro default = 30 days. Enterprise = custom. Regulatory requirements (e.g. MAS in Singapore) may drive longer retention.
8Are you currently using Elasticsearch/Kibana for log search? What daily ingest volume?
Sizes the ELK replacement opportunity. Loki is typically 50–75% cheaper than ELK at equivalent volume.

Users & General

9How many users need access to Grafana dashboards?
Cloud Pro: 3 included, then $8/additional active user/month (or $55 with all Enterprise plugins).
10Do you need SSO / SAML / LDAP integration?
Included in Cloud (all tiers). Requires Enterprise license for self-managed. Impacts deployment choice.
11Are there data residency or sovereignty requirements?
Singapore regions available on AWS + GCP. For strict air-gap or in-country-only: route to BYOC or self-managed Enterprise.
12How do you expect usage to grow over 6–12 months?
Helps size an annual commit (Enterprise tier) vs. pay-as-you-go (Pro). Enterprise pricing starts at $25K/year with volume discounts.
💰 Quick Reference — Cloud Pro List Rates
Metrics: $6.50 / 1K active series / month (at 1 DPM)  |  Logs: ~$0.55/GB (30-day retention)  |  Users: $8/active user/month  |  Platform fee: $19/month
Enterprise: custom pricing from $25K/year minimum, as low as $3/1K series at volume. See grafana.com/pricing.

3   Grafana OSS vs Enterprise vs Cloud

Three deployment options — same core visualization engine, different levels of management and capability.

Capability Grafana OSS
Free, self-managed		 Grafana Enterprise
Licensed, self-managed		 Grafana Cloud
Fully managed SaaS
Hosting Customer manages everything Customer manages everything Grafana Labs manages — HA, upgrades, backups automatic
License AGPLv3 — free to use Commit-based. Base price includes 1 production instance + 60 users Consumption-based (pay for active users + telemetry ingested)
Data Sources 60+ OSS plugins 60+ OSS + 20+ Enterprise plugins (SAP HANA, Oracle, Snowflake, ServiceNow, etc.) All OSS + all Enterprise plugins included
LGTM Backends Self-hosted Mimir, Loki, Tempo, Pyroscope Self-hosted (was GEM/GEL/GET — deprecated 2025; use OSS backends) Managed Mimir, Loki, Tempo, Pyroscope — no infrastructure to run
SSO / SAML / LDAP ✘ Not included ✔ Included ✔ Included (all tiers)
RBAC Basic (Org-level roles) ✔ Fine-grained ✔ Fine-grained
Reporting & PDF ✘ Not included ✔ Included ✔ Included
Query Caching ✘ Not included ✔ Redis / Memcached ✔ Built-in
Infra Integrations (100+) ✘ Self-assembled ✘ Self-assembled ✔ Turnkey catalog
Adaptive Telemetry ✘ ✘ ✔ Metrics, Logs, Traces
Incident Response (IRM) ✘ Partial (OnCall OSS) ✔ Full IRM suite
Synthetic Monitoring ✘ ✘ ✔ 22 global probes incl. Singapore
AI / Assistant ✘ ✘ ✔ Grafana Assistant
Support Community only Enterprise support + SLA 8×5 (Pro), 24×7 custom SLA (Enterprise)
Upgrades Manual — you own it Manual — you own it Automatic — always latest

🎯 Recommendation for This Environment

Grafana Cloud is the strongest fit for this stack because:

  • Breadth: 6 of 10 services get immediate turnkey or JVM-level coverage. No exporter assembly.
  • ELK replacement: Loki replaces Elasticsearch for log search at 50–75% lower cost; Grafana replaces Kibana for visualization.
  • No host tax: Pay for metrics ingested, not per host — significant savings at this service count.
  • Managed backends: No Mimir/Loki/Tempo infrastructure to stand up or maintain.
  • Singapore availability: Both AWS and GCP regions available (see below).

Grafana Enterprise (self-managed) is the fallback if strict air-gap or on-prem-only requirements exist. BYOC is the middle ground — Grafana-managed in the customer's own cloud VPC.

4   Grafana Cloud — Singapore Availability

Grafana Cloud has 27 regions across AWS, GCP, and Azure. Singapore is available on two providers.

ProviderRegion CodeAvailabilityHow to Access
AWS ap-southeast-1 ✔ Available Grafana Self-Serve signup + AWS Marketplace
GCP asia-southeast1 ✔ Available Google Cloud Marketplace
Azure — ✘ Not in Singapore Nearest Azure: Netherlands (westeurope) or US Central (centralus)

📍 What This Means for the Customer

  • Data stays in Singapore: Stacks created in ap-southeast-1 (AWS) or asia-southeast1 (GCP) keep data region-scoped. Note: Grafana does not make a formal in-region data-residency guarantee — confirm with the account team before committing if the customer has strict MAS/PDPA requirements.
  • Synthetic Monitoring probe: A public probe runs in Singapore (AWS) for latency checks from within the region.
  • Support: Grafana has APAC support team members based in Singapore, Australia, India, and Indonesia.
  • If air-gap or strict on-prem is required: Route to BYOC (Grafana-managed in the customer's own AWS/GCP VPC) or Grafana Enterprise (fully self-managed).
⚠️ Region lock: Once a Grafana Cloud stack is created in a specific region, it cannot be migrated to another region. Choose the region at setup carefully. Verify current availability at grafana.com regional availability docs.

5   Other APAC Regions (Reference)

If the customer has infrastructure outside Singapore.

LocationAWSGCP
Australia✔ ap-southeast-2✔ australia-southeast1
India✔ ap-south-1✔ asia-south1
Indonesia✔ ap-southeast-3—
Japan✔ ap-northeast-1—
Grafana Labs  ·  Prepared by Rene Wong, Strategic Channel Manager  ·  September 2026
Integration catalog snapshot: 2026-05-19. Always verify against grafana.com/integrations and grafana.com/pricing before pitching.