Getting Started¶
Intel® ESQ runs on a range of supported platforms with discrete GPU and NPU accelerators. The architecture adapts to the available hardware, exercising compute resources through a structured stack of runtimes, test suites, and application-level workloads.
%%{init: {'theme': 'base', 'themeVariables': {'fontSize': '14px', 'primaryColor': '#F5F5F5', 'primaryBorderColor': '#BDBDBD', 'lineColor': '#757575', 'edgeLabelBackground': '#FFFFFF'}}}%%
block
columns 6
A_OV["OpenVINO*"] A_DLS["DL Streamer*"] A_OVMS["OVMS*"] A_VLLM["vLLM*"] A_APP1["STREAM*"] S_OTH_APP["App-N"]
block:T_BLK:6
columns 7
space:3 T_HDR["Test Cases"] space:3
end
S_AI["AI"] S_VSN["Vision"] S_SYS["System"] S_RT["Real-Time"] S_VRT["Vertical"] S_OTH["Test Suite-N"]
block:S_BLK:6
columns 7
space:3 S_HDR["Test Suites"] space:3
end
block:CORE_BLK:6
columns 5
space:2 CORE_HDR["CLI"] space:2
end
block:RUN_BLK:6
columns 5
space:2 RUN_HDR["Runtimes"] space:2
space RUN_DK["Docker*"] RUN_PY["Python*"] RUN_ND["Node.js*"] space
end
block:OS_BLK:6
columns 5
space:2 OS_HDR["Operating system"] space:2
space:2 OS_UBUNTU["Ubuntu*"] space:2
end
classDef whiteNodeDash fill:#f0f0f0,stroke:#A0A0A0,stroke-dasharray: 4;
classDef greenNode fill:#C9EED5,stroke:#A0A0A0,stroke-width:0.2px;
classDef blueNode fill:#C7DAE9,stroke:#A0A0A0,stroke-width:0.2px;
classDef blueHdrNode fill:#C7DAE9,stroke:#C7DAE9;
classDef blueBlkNode fill:#C7DAE9,stroke:#A0A0A0,stroke-width:0.2px;
classDef greyHdrNode fill:#f0f0f0,stroke:#f0f0f0;
classDef greyBlkNode fill:#f0f0f0,stroke:#A0A0A0,stroke-width:0.2px;
class A_OV,A_DLS,A_OVMS,A_VLLM,A_APP1,A_APP2 greenNode
class S_AI,S_VSN,S_SYS,S_RT,S_VRT blueNode
class OS_UBUNTU,RUN_PY,RUN_DK,RUN_ND greenNode
class CORE_HDR,S_HDR,T_HDR blueHdrNode
class S_OTH,S_OTH_APP whiteNodeDash
class CORE_BLK,S_BLK,T_BLK blueBlkNode
class OS_BLK,RUN_BLK greyBlkNode
class OS_HDR,RUN_HDR greyHdrNode
The following table describes each component in the stack, from the application layer down to the operating system foundation.
| Layer | Description |
|---|---|
| Applications and Components | Workloads and services exercised by test cases — OpenVINO™ toolkit, Intel® DL Streamer*, OVMS*, vLLM*, and custom applications |
| Test Cases | Individual parameterized test cases within each suite, each targeting a specific workload or use case |
| Test Suites | Domain-specific test suites — AI, Vision, System, Real-Time, Vertical, and extensible custom suite types |
| CLI | Intel® ESQ command-line interface — the primary entry point for running tests, managing profiles, and viewing results |
| Runtimes | Docker*, Python*, and Node.js* runtime environments required for test execution and containerized workloads |
| Operating System | Ubuntu* as the supported base Linux* distribution |
Next Steps¶
- Quick Start – Install dependencies and run all tests.
Need Help?¶
If you encounter issues during setup:
- Refer to the Optimization guide.
- Check the Troubleshooting guide.
- Visit GitHub* Issues for community support.