阅读中文版本

Build a Skip List Index

Skip List Index — a complete walkthrough of ordered in-memory indexing, covering level generator, node inserter, search path, and testing.

skip_list_index/
├── src/index.ts
├── src/level-generator.ts
├── src/node-inserter.ts
├── tests/skip_list_index.test.ts
└── package.json

// Level generator: ordered in-memory indexing core
export function levelgenerator(input: LevelgeneratorInput): LevelgeneratorResult {
  // Build the level generator that anchors ordered in-memory indexing and defines the core data shape.
  return processLevelgenerator(input);
}

// Node inserter — handles ordered in-memory indexing Add the node inserter so
export class Nodeinserter {
  constructor(private opts: NodeinserterOptions) {}
  async run(ctx: Context): Promise<NodeinserterOutput> {
    return this.execute(ctx);
  }
}

// Search path: ordered in-memory indexing stage
export const searchpath = (next: Next) => async (req: Request) => {
  const out = await transform(req);
  return next(out);
};

// Range scanner for ordered in-memory indexing
export async function rangescanner(stream: AsyncIterable<Chunk>): Promise<Aggregate> {
  let acc = init();
  for await (const c of stream) acc = merge(acc, c);
  return acc;
}

// Delete marker: ordered in-memory indexing finalization
export function deletemarker(state: State): Result {
  validate(state);
  return finalize(state);
}

Project overview

This project builds a ordered in-memory indexing system whose reason for existing is maintaining a sorted structure with logarithmic insert, delete, and range queries without rebalance cost. The temptation in ordered in-memory indexing is to bolt the interesting part onto a working prototype and discover the hard cases in production; this build inverts that order and treats a degenerate promotion sequence that produces tall, sparse levels and degrades search toward linear as the first thing designed, not the last.

You will implement Level generator, Node inserter, Search path, Range scanner, and Delete marker, each as a module with one responsibility and a test that proves it in isolation. The ordered in-memory indexing pipeline is then the wiring that composes them, and the integration test is where the composition is shown to survive a degenerate promotion sequence that produces tall, sparse levels and degrades search toward linear.

The walkthrough is written for a developer who has used a ordered in-memory indexing system but has not had to make one honest. It favors explicit data shapes over clever abstractions, and it names the failure each design choice prevents rather than presenting the choice as a style preference.

Each phase ends with something observable — a level generator that returns the right shape, a node inserter that rejects bad input, an integration test that goes green — so you can stop at any point and still have a ordered in-memory indexing system that works for the cases covered so far.

What you will build

The finished ordered in-memory indexing system exposes five capabilities, each with a focused test:

• Level generator
• Node inserter
• Search path
• Range scanner
• Delete marker

Level generator is the foundation that every later stage assumes; Node inserter is where ordered in-memory indexing correctness lives; Search path handles a degenerate promotion sequence that produces tall, sparse levels and degrades search toward linear; Range scanner makes the system observable; and Delete marker is the graceful-degradation layer that keeps overload from becoming corruption.

The build keeps the ordered in-memory indexing logic in ordinary modules you can read in one sitting, and pulls in dependencies only for the parts that are genuinely tedious to hand-roll. That keeps the diff between the tutorial and a real codebase small, and keeps the failure modes visible in code rather than hidden behind a framework.

Architecture and data model

The system is organized around Level generator, Node inserter, and Search path, with Range scanner tying them together. Data flows as a multi-level linked list where each level skips over more nodes, giving expected logarithmic search through random promotion.

Each component is a module with a pure interface where practical: the same input produces the same output, and side effects are pushed to the edges. That discipline is what makes the phases independently testable, because a test for Search path does not need to stand up Level generator and Node inserter to exercise the interesting branch.

The data model is deliberately explicit. A multi-level linked list where each level skips over more nodes, giving expected logarithmic search through random promotion Keeping the shape written down — in a schema, a type, or a fixture — means a change to the shape is a visible diff rather than a silent runtime shift, and it means the integration test can assert against a concrete shape instead of guessing.

Configuration and code are kept apart: the configuration says what ordered in-memory indexing should do, the modules say how. A reviewer can change a threshold or a policy without reading the implementation, which keeps the risky changes small and lets the diff stand as the review.

The failure path through Search path is designed before the happy path is finished, because a degenerate promotion sequence that produces tall, sparse levels and degrades search toward linear is the case that distinguishes a ordered in-memory indexing system from a demo. Every component declares what it does when its inputs are wrong, and the integration test asserts each of those declarations.

Prerequisites and dependencies

Have these ready before you install anything, so the ordered in-memory indexing setup is not blocked by a missing tool:

• Node.js 20 LTS or later
• Familiarity with ordered in-memory indexing fundamentals
• A local environment to run the test suite

The runtime dependencies are chosen to keep the ordered in-memory indexing datapath legible:

• fastify (^5.0.0) — Index API
• pino (^9.0.0) — Operation logging

# setup for skip-list-index
pnpm init
pnpm add fastify pino
pnpm add -D typescript tsx vitest

Versions are pinned to a major range so the ordered in-memory indexing install is reproducible; commit the lockfile. The level generator reads its operational parameters from the environment and validates them at startup, so a missing value fails the boot rather than the first ordered in-memory indexing request.

Directory structure

Keep each ordered in-memory indexing component in its own file so it can be tested without starting the others:

skip_list_index/
├── src/index.ts
├── src/level-generator.ts
├── src/node-inserter.ts
├── tests/skip_list_index.test.ts
└── package.json

Phase 1: Level generator

Build the level generator that anchors ordered in-memory indexing and defines the core data shape.

// Level generator: ordered in-memory indexing core
export function levelgenerator(input: LevelgeneratorInput): LevelgeneratorResult {
  // Build the level generator that anchors ordered in-memory indexing and defines the core data shape.
  return processLevelgenerator(input);
}

After this phase the level generator works in isolation and is covered by a unit test. The level generator defines the data shape that every later stage assumes, so getting it right first means the later phases compile against a stable contract.

Treat the level generator as the single source of truth for the ordered in-memory indexing data shape. Validate inputs at the boundary and reject early, so a malformed value never reaches Node inserter or Search path where its effect would be harder to localize.

Phase 2: Node inserter

Add the node inserter so ordered in-memory indexing can maintaining a sorted structure correctly.

// Node inserter — handles ordered in-memory indexing Add the node inserter so
export class Nodeinserter {
  constructor(private opts: NodeinserterOptions) {}
  async run(ctx: Context): Promise<NodeinserterOutput> {
    return this.execute(ctx);
  }
}

After this phase the node inserter works in isolation and is covered by a unit test. This is where ordered in-memory indexing correctness lives: the node inserter is what makes maintaining a sorted structure with logarithmic insert, delete, and range queries without rebalance cost true, and its test is the proof.

Keep the node inserter pure where you can. A function that takes an input and returns an output is trivial to test and impossible to mock wrong; a class with hidden state is the opposite. Push side effects to the caller and keep the node inserter about the transformation.

Phase 3: Search path

Wire the search path to handle a degenerate promotion sequence without losing data.

// Search path: ordered in-memory indexing stage
export const searchpath = (next: Next) => async (req: Request) => {
  const out = await transform(req);
  return next(out);
};

After this phase the search path works in isolation and is covered by a unit test. The search path is the difference between a ordered in-memory indexing system that works on the happy path and one that survives a degenerate promotion sequence that produces tall, sparse levels and degrades search toward linear.

Design the failure before the success. Decide what search path does when a degenerate promotion sequence that produces tall, sparse levels and degrades search toward linear: reject, retry, shed, or degrade — and write that test first. A ordered in-memory indexing system whose failure behavior is unspecified will pick one at random under pressure, and it is usually the wrong one.

Phase 4: Range scanner

Integrate the range scanner to make ordered in-memory indexing observable and testable end to end.

// Range scanner for ordered in-memory indexing
export async function rangescanner(stream: AsyncIterable<Chunk>): Promise<Aggregate> {
  let acc = init();
  for await (const c of stream) acc = merge(acc, c);
  return acc;
}

After this phase the range scanner works in isolation and is covered by a unit test. Observability is not a finishing touch; it is how you know the ordered in-memory indexing system is doing what it claims in production.

Emit one structured event per meaningful action, with enough context to localize a failure without a second round-trip. The range scanner is what turns “something is slow” into “the node inserter is slow for inputs matching this shape,” which is the difference between an alert and an investigation.

Phase 5: Delete marker

Finalize the delete marker so the ordered in-memory indexing system degrades gracefully under overload.

// Delete marker: ordered in-memory indexing finalization
export function deletemarker(state: State): Result {
  validate(state);
  return finalize(state);
}

After this phase the delete marker works in isolation and is covered by a unit test. The delete marker is what keeps overload from becoming corruption: when the ordered in-memory indexing system is asked for more than it can do, it degrades in a defined way instead of failing in an undefined one.

Pick a degradation strategy for delete marker — shed, queue, or backpressure — and make it explicit in code and in a test. Leaving the behavior to chance is how a ordered in-memory indexing traffic spike becomes a data loss incident.

Testing

Each phase has a focused test: level generator, node inserter, and search path are exercised in isolation, then an integration test drives the full ordered in-memory indexing pipeline and asserts the a degenerate promotion sequence case is handled.

Start with unit tests that pin each component’s contract: the level generator returns the documented shape, the node inserter transforms correctly, and the search path fails in the documented way. These tests are fast and they localize regressions to the component that broke.

Then add an integration test that composes the components and drives a realistic ordered in-memory indexing scenario end to end. Assert the happy path first, then deliberately trigger a degenerate promotion sequence that produces tall, sparse levels and degrades search toward linear and assert the system degrades rather than corrupts. Because the components are pure where practical, the integration test stays deterministic rather than flaky.

A ordered in-memory indexing suite must also prove the negative: bad input is rejected at the boundary, overload is shed rather than queued indefinitely, and a partial failure does not leave the system in an inconsistent state. These are the tests that catch the regressions a happy-path suite never would.

The entire ordered in-memory indexing suite runs with no external service; every dependency is either in-process or stubbed, so every case is reproducible in CI. That is what lets the search path failure-mode assertions live as tests rather than as a manual checklist that drifts out of date.

# run the ordered in-memory indexing test suite
pnpm vitest run

Deployment and operations

Run at least two ordered in-memory indexing instances behind a load balancer and share state in Redis. Set resource limits so ordered in-memory indexing overload surfaces as backpressure rather than failure.

Run at least two stateless instances behind a load balancer, and move any shared state into a shared store so the ordered in-memory indexing behavior is global rather than per instance. Size limits so the system sheds load before it runs out of memory, because a ordered in-memory indexing system that OOMs under load fails in a way that is hard to reason about.

Drain on deploy: stop accepting new work, let in-flight ordered in-memory indexing operations finish, then exit. Because the system holds no state that cannot be reconstructed, a rollout is a matter of cycling instances one at a time while watching the error rate.

When a dependency is unavailable, decide explicitly whether the ordered in-memory indexing system fails closed or degrades. Leaving that choice to chance is how a dependency outage becomes a silent data loss, so make the choice visible in code and in a test.

Security considerations

Validate all ordered in-memory indexing inputs at the boundary, reject oversized payloads before buffering them, and keep secrets in the environment rather than the repository.

The ordered in-memory indexing system is a choke point, so it is the right place to reject malformed input before it reaches node inserter or search path. Validate the shape and the size at the edge, and let the inner components assume what they receive is well-formed.

Never log secrets or full payloads; log enough to localize a failure without recording data that should not be on disk. Treat any external call as untrusted and validate its response before using it, because a ordered in-memory indexing system that trusts its inputs blindly is one malformed response away from corruption.

Keep dependencies pinned and run an audit in CI. A transitive dependency with a known vulnerability is still a vulnerability in your ordered in-memory indexing system, and the lockfile is what makes the audit meaningful.

Observability

Emit one structured event per ordered in-memory indexing operation with the inputs’ shape, the outcome, and a correlation id. Track error rate and latency per component and alert on a step change rather than a single spike.

The correlation id is the thread that ties a user-visible symptom to the ordered in-memory indexing decision and to the component log entry for the same operation. Without it, an investigation starts from a timestamp and a guess.

Watch the search path failure rate separately from the overall error rate. A jump in search path failures usually means a degenerate promotion sequence that produces tall, sparse levels and degrades search toward linear, not a bug in node inserter, and the two have different owners and different fixes.

Chart the share of ordered in-memory indexing operations that shed, queue, or succeed. A change in the mix points at the cause faster than an aggregate latency can, because each class has a different root.

Troubleshooting

A degenerate promotion sequence that produces tall, sparse levels and degrades search toward linear. Reproduce with a minimal fixture, then add a guard test so the ordered in-memory indexing system rejects the condition loudly.

Performance regresses under ordered in-memory indexing load. Profile with the sampling analyzer, locate the hot path in node inserter, and add a targeted benchmark to lock the regression.

Keep the ordered in-memory indexing dashboards separate from dependency dashboards. When an operation fails, the correlation id lets you decide in seconds whether the fault is level generator, node inserter, or search path, instead of guessing from a timeout.

Limitations and next steps

• Single-region; multi-region replication is out of scope.
• The level generator is in-process and not horizontally shared by default.
• No auth layer; assume deployment behind an edge that enforces identity.

The ordered in-memory indexing system is single-region by design; extending to multi-region means making level generator shareable across regions, which is a real change to the consistency model rather than a config flag. The search path is in-process, which is correct for one instance but must move to a shared store before scaling out.

There is no authentication or authorization layer; the system assumes it is deployed behind an edge that enforces identity. Adding one is straightforward but out of scope for this build, and doing it here would distract from the ordered in-memory indexing correctness that is the point.

The natural next step, once the ordered in-memory indexing pipeline is trusted in production, is to add a short-TTL cache for idempotent reads and a richer range scanner that exports histograms rather than counters. Both are additive and neither changes the contracts this build establishes.