Live Demo
1 / 15
Database Performance, Security, & Management Framework
847
Knowledge Nodes
91%
Confidence Score
3
Active Alerts
14
Days Learning
Self-learning database intelligence for SQL Server. It doesn't just monitor your databases — it studies them.
Sentinel Dashboard
Real-Time Monitoring
Live across 4 monitored instances
847
KG Nodes
+12 today
91%
Confidence
Improving
3
Active Alerts
1 critical
18.4s
Top Query
340% above
AlertInstanceValueStatus
Query 0x4A3F spike — OrdersDBPROD-0118.4sCRITICAL
CXPacket waits elevatedREPORT-0247msWARNING
Missing index detected — OrdersPROD-01Impact 94%INFO
CPU — PROD-01
Memory PLE
Query Intelligence
Query Performance Analysis
Ranked by deviation from learned baseline
Query HashDatabaseAvg msBaselineExecutionsStatus
0x4A3F...OrdersDB18,421400ms847+340%
0xBF92...ReportingDB2,8401,900ms312+49%
0x3D11...OrdersDB145155ms4,821Normal
0xA271...AppDB8891ms12,440Normal
Executions today — 0x4A3F
Avg duration trend
Query Intelligence
Root Cause Analysis
AI-generated for query 0x4A3F — OrdersDB
⚠ Root Cause Identified
Primary: Missing non-clustered index on Orders.CustomerID + OrderDate
Secondary: Parameter sniffing — plan cached for VIP customer with 48,000 orders
Correlated: Bulk import of 2.3M rows at 13:58 triggered cache flush at 14:05
Recommended fix — estimated 94% cost reduction:
CREATE NONCLUSTERED INDEX IX_Orders_CustomerID_Date
ON dbo.Orders (CustomerID, OrderDate)
INCLUDE (OrderTotal, Status, ShipDate);
-- Est. cost reduction: 94% | Affected queries: 847/day
Impact estimate
Duration reduction94%
Read reduction91%
Affected queries/day
847
executions will benefit
Knowledge Graph
Self-Learning Intelligence
847 nodes — 14 days of continuous learning
847
Total Nodes
91%
Confidence
4,821
Edges
94%
FP Reduction
Node Types in Graph
312
Query Hashes
228
Wait Types
307
Resource Events
Learning Timeline
Days 1–7: Silent observation
Baselines built, no alerts fire
Day 8: Alerting begins
3σ deviations trigger intelligent alerts
Day 30: <6% false positive rate
Knowledge graph matures, patterns confirmed
AI Assistant
Sentinel AI Chat
Natural language database intelligence
Why is OrdersDB slow right now?
Sentinel: Query 0x4A3F is running at 18.4s — 340% above its 14-day baseline of 400ms. Root cause: missing index on Orders.CustomerID combined with a parameter sniff from a VIP customer cache. This was preceded by a 2.3M-row bulk import at 13:58, which flushed the plan cache 7 minutes later. The correlated event chain is: import → cache flush → suboptimal plan → spike.
Generate the fix script
Sentinel: Here's the remediation script:
CREATE NONCLUSTERED INDEX IX_Orders_CustomerID_Date
ON dbo.Orders (CustomerID, OrderDate)
INCLUDE (OrderTotal, Status);

EXEC sp_query_store_force_plan @query_id=4291, @plan_id=2847;
-- Forces previous good plan while index is built
Wait Analysis
Wait Statistics
Current vs 7-day learned baseline
Wait TypeCurrentBaselineTrendStatus
CXPACKET47ms12ms
+292%
LCK_M_S38ms4ms
+850%
PAGEIOLATCH_SH8ms9ms
Normal
AI Interpretation
CXPACKET: Parallel query contention. Correlated to ReportingDB parallel jobs running 14:00–15:00 daily — DPSMF has learned this pattern and will auto-suppress from tomorrow.

LCK_M_S: Lock contention directly linked to the missing index on Orders. Resolves when index is created.
Anomaly Detection
Anomaly Timeline
Causal chain discovered automatically
🔗 Correlated Event Chain — Today
09:17
Parameter sniff plan change
Execution plan for 0x4A3F swapped to suboptimal variant
13:58
Bulk insert — 2.3M rows
Orders table — triggered plan cache flush 7 minutes later
14:05
Plan cache flush
Suboptimal plan recompiled without parameter optimization
14:23
Query spike — 18.4s (ALERT)
0x4A3F duration 340% above baseline
14:25
DPSMF correlates chain — root cause identified
3 events linked. Fix script generated automatically.
🤖 DPSMF connected these 3 events in real time. Traditional monitoring tools would have shown 3 separate unrelated alerts. DPSMF built the complete causal narrative automatically.
Configuration Audit
Config Audit Grades
Best-practice analysis across all instances
A
AppDB
All checks passed
B
OrdersDB
2 warnings
C
ReportingDB
MAXDOP issue
CheckFindingGrade
MAXDOP Configuration0 (unlimited)C
Query Store StatusEnabledA
TempDB Files2 (rec: 8)B
Backup ScheduleDaily + logA
Index Intelligence
Index Advisor
AI-generated recommendations ranked by impact
RecommendationImpactAffected/DayAction
CREATE IX on Orders.CustomerID + OrderDate94%847 queriesHigh
Add INCLUDE cols to IX_Orders_Status67%312 queriesMed
Remove duplicate IX_Cust_Email (unused)240MB savedSpace onlyLow
Rebuild IX_Orders_Date (82% fragmented)12%All readsMaint
Top recommendation impact
Duration -94%18.4s → 1.1s
Reads -91%2.1M → 190K
Total daily queries to benefit
1,159
across 4 recommendations
Security Audit
Security Assessment
Continuous monitoring across all instances
A-
Overall Grade
14
Passed
3
Warnings
1
Critical
FindingInstanceSeverity
SA account — weak password policyPROD-01Critical
xp_cmdshell enabledPROD-01Warning
TDE not active on dev serverDEV-03Warning
Guest account has DB accessREPORT-02Warning
🔒 Security posture is monitored continuously — any change triggers an immediate alert. DPSMF detected the SA password policy issue 3 days after the server was provisioned.
Plan Cache
Plan Cache Intelligence
Execution plan regression detection
2,847
Cached Plans
94%
Hit Ratio
3
Regressions
1
Plans Forced
QueryChangeImpactAction
0x4A3F09:17+340%Force plan?
0xF211Yesterday+48%Monitoring
0x8C143 days agoStableForced ✓
📄 DPSMF can force the previous known-good plan for 0x4A3F automatically when Query Store is enabled — restoring normal performance in seconds without a code change.
Capacity Planner
Capacity & Growth Forecasts
Projected constraint dates at current growth rate
InstanceMetricCurrentProjected FullStatus
PROD-01Data volume (D:)78%12 daysCritical
REPORT-02Transaction log61%24 daysWarning
PROD-01Connection count42%> 90 daysOK
PROD-01 data growth (30 days)
R²=0.97 — high confidence projection
Recommended action
Add 500GB to D: drive on PROD-01 within 7 days. At current growth rate (12GB/day), threshold will be reached Apr 8.
Operations
Reporting & Instance Management
Fleet-wide oversight from a single dashboard
Scheduled Reports
Daily Performance SummaryDaily 06:00
Query Regression ReportWeekly Mon
Security Audit SummaryWeekly Fri
Capacity Planning ForecastWeekly Mon
Instance Fleet
PROD-011 Critical
REPORT-021 Warning
DEV-03Healthy
STAGING-04Healthy
Availability Group — PROD-AG1
Primary
PROD-01 ✓
Secondary
DR-05 Sync
Redo Queue
0 KB
Demo Complete
That's DPSMF.
In this demo, DPSMF identified a critical regression, traced the root cause across a 26-minute event chain, generated a precise fix, assessed security, graded configuration, planned capacity, and produced a prioritised action list — all automatically.
🔍
Query spike traced to bulk import 26 min prior
Causal chain discovered automatically
🎯
Missing index — 94% cost reduction fix generated
Exact CREATE INDEX statement, ready to execute
🔒
Security grade A- — 1 critical finding surfaced
Continuous posture monitoring, no manual review needed
📈
Storage critical in 12 days — action recommended
R²=0.97 projection confidence
Start Free Trial → View User Guide
Act 1 — Platform
Welcome to DPSMF
Platform overview