Inference Systems · Cluster
Graph Compilation
Fusion, scheduling, and kernels.
- 01 Computational Graphs as an Intermediate Representation Why a model compiles to a list of pure value-producing nodes, how topological order doubles as the schedule, and which properties of a program a graph cannot express.
- 02 Constant Folding and Algebraic Rewrites on a Fixed Graph How a compiler evaluates constant subgraphs at build time, which algebraic identities survive floating-point arithmetic, and why rewrite order determines the result.
- 03 Operator Fusion and the Conditions That Block It Why merging adjacent operators into one kernel removes memory traffic, how producers and consumers are classified, and which graph shapes make the merge illegal or unprofitable.
- 04 Static Memory Planning from Tensor Liveness Intervals Turning a liveness interval per tensor into an offset in one arena, why interval-graph colouring is optimal for monotone lifetimes, and where greedy packing loses bytes.
- 05 Layout Assignment and the Elimination of Transposes A tensor has one logical shape and many physical orders; choosing one per value turns transpose chains into bookkeeping and decides which kernels can read their inputs directly.
- 06 Dynamic Shapes, Guards, and the Cost of Recompilation A compiled kernel needs concrete extents; symbolic shapes force either a general kernel, a bucket, or a recompile, and each choice moves cost to a different place.
- 07 Tile Selection: Cost Models Against Measurement A fused kernel still needs a tile shape and a loop order, and the compiler must choose them either by estimating cost or by measuring candidates on the device.
- 08 Kernel Launch Capture and Replay Recording a launch sequence once and replaying it as a single submission removes per-launch host overhead, at the cost of fixed topology and a separate memory model.
- 09 Inter-Operator Parallelism and Stream Scheduling Independent branches of a graph can run concurrently on separate queues, but overlap only helps when the branches use different resources and the dependencies are exact.
- 10 Graph Breaks, Fallback Paths, and Numerical Drift What happens at the boundary where tracing stops, why the eager fallback does not produce bit-identical results, and how a runtime keeps two execution paths consistent.