docs: add back-pressure and cancellation blog (#479)

* docs: add back-pressure and cancellation blog

* fix(doc): format and content improvement for the back-pressure blog (#496)

* Minor editorial changes on cancellation blog.

* adjust blog structure

* rename blog title

---------

Co-authored-by: Wang Zixiao <wayne.wang0821@gmail.com>
Co-authored-by: Meng Zhang <meng@tabbyml.com>

website/blog/2023-09-30-stream-laziness-in-tabby/cancellation.png

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93f559473aa7d0e2da81d2fb2fe2a733549a2b83cc84da8c95d92d0bc4d2e8b5
+size 130476

website/blog/2023-09-30-stream-laziness-in-tabby/index.md

@@ -0,0 +1,139 @@
+---
+authors: [wwayne, gyxlucy, meng]
+
+tags: [tech design]
+---
+
+# Stream laziness in Tabby
+
+This blog focuses on understanding stream laziness in Tabby. You do not need to know this information to use the Tabby, but for those interested, it offers a deeper dive on why and how the Tabby handle its LLM workload.
+
+## What is streaming?
+Let's begin by setting up a simple example program:
+
+![intro](./intro.png)
+
+```javascript
+const express = require('express');
+
+function sleep(ms) {
+  return new Promise(resolve => setTimeout(resolve, ms));
+}
+
+async function* llm() {
+  let i = 1;
+  while (true) {
+    console.log(`producing ${i}`);
+    yield i++;
+
+    // Mimic LLM inference latency.
+    await sleep(1000);
+  }
+}
+
+function server(llm) {
+  const app = express();
+  app.get('/', async (req, res) => {
+    res.writeHead(200, {
+      'Content-Type': 'application/jsonstream',
+      'Transfer-Encoding': 'chunked',
+    });
+
+    let value, done;
+    do {
+      ({ value, done } = await llm.next());
+      res.write(JSON.stringify(value));
+      res.write('\n');
+    } while (!done);
+  });
+
+  app.listen(8080);
+}
+
+async function consumer() {
+  const resp = await fetch('http://localhost:8080');
+
+  // Read values from our stream
+  const reader = resp.body.pipeThrough(new TextDecoderStream()).getReader();
+  // We're only reading 3 items this time:
+  for (let i = 0; i < 3; i++) {
+    // we know our stream is infinite, so there's no need to check `done`.
+    const { value } = await reader.read();
+    console.log(`read ${value}`);
+  }
+}
+
+server(llm());
+consumer();
+```
+
+## Stream Laziness
+
+If you were to run this program, you'd notice something interesting. We'll observe the LLM continuing to output `producing ${i}` even after the consumer has finished reading three times. This might seem obvious, given that the LLM is generating an infinite stream of integers. However, it represents a problem: our server must maintain an ever-expanding queue of items that have been pushed in but not pulled out.
+
+Moreover, the workload involved in creating the stream is typically both expensive and time-consuming, such as computation workload on the GPU. But what if the client aborts the in-flight request due to a network issue or other intended behaviors?
+
+This is where the concept of stream laziness comes into play. We should perform computations only when the client requests them. If the client no longer needs a response, we should halt production and pause the stream, thereby saving valuable GPU resources.
+
+![Cancellation](./cancellation.png)
+
+## How to handle canellation?
+
+The core idea is straightforward: on the server side, we need to listen to the `close` event and check if the connection is still valid before pulling data from the LLM stream.
+
+```js
+app.get('/', async (req, res) => {
+  ...
+  let canceled;
+  req.on('close', () => canceled = true);
+  do {
+    ({ value, done } = await llm.next());
+    ...
+  } while (!done && !canceled);
+});
+```
+
+## Implement cancellation for Tabby
+
+In Tabby, effective management of code completion cancellations is crucial for promptly responding to users' inputs while optimizing model usage to enhance performance.
+
+On the client side, whenever we receive a new input from a user, it's essential to abort the previous query and promptly retrieve a new response from the server.
+
+```js
+// Demo code in the client side
+
+let controller;
+
+const callServer = (prompt) => {
+  controller = new AbortController();
+  const signal = controller.signal;
+  // 2. calling server API to get the result with the prompt
+  const response = await fetch("/v1/completions", {
+    method: "POST",
+    headers: {
+      "Content-Type": "application/json"
+    },
+    body: JSON.stringify({ prompt })
+    signal
+  });
+}
+
+const onChange = (e) => {
+  if (controller) controller.abort(); // Abort the previous request
+  callServer(e.target.value);
+};
+
+// 1. Debounce the input 100ms for example
+<input onChange={debounce(onChange)} />
+
+```
+
+By employing streaming and implementing laziness semantics appropriately, all components operate smoothly and efficiently!
+
+![Streaming](./stream.png)
+
+## That's it
+
+We would love to invite to join our Slack community! Please feel free to reach out to us on [Slack](https://join.slack.com/t/tabbycommunity/shared_invite/zt-1xeiddizp-bciR2RtFTaJ37RBxr8VxpA) - we have channels for discussing all aspects of the product and tech, and everyone is welcome to join the conversation.
+
+Happy hacking 😁💪🏻

website/blog/2023-09-30-stream-laziness-in-tabby/intro.png

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4113704a91e8355fb14ff43c0b26fb3fd926168dbc0eacca008f5b8657e1cb8e
+size 111767

website/blog/2023-09-30-stream-laziness-in-tabby/stream-flush.png

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:905ca358ae3198ccd587d7a837cc82fcbfad30c7c5b6115248a41ed9ae5468b8
+size 107185

website/blog/2023-09-30-stream-laziness-in-tabby/stream.png

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74734640c70fb1ebbfe6baa9a3ff8f8ef257577766d98df6bbdb066e44a9a17d
+size 104103

website/blog/authors.yml

@@ -7,3 +7,13 @@ gyxlucy:
   name: Lucy Gao
   url: https://github.com/gyxlucy
   image_url: https://github.com/gyxlucy.png
+
+icycodes:
+  name: Zhiming Ma
+  url: https://github.com/icycodes
+  image_url: https://github.com/icycodes.png
+
+wwayne:
+  name: Wayne Wang
+  url: https://github.com/wwayne
+  image_url: https://github.com/wwayne.png