Skip to content

Latest commit

 

History

History
76 lines (54 loc) · 4.69 KB

HOW_TO_RUN_DEMOS.md

File metadata and controls

76 lines (54 loc) · 4.69 KB
Raw

node-redis-queue-demos

Examples of how to use node-redis-queue.

##Running the demos -- preliminary steps

  1. Open two Git Bash console windows.
  2. In each window run cd demo/lib and then export NODE_PATH=../../...
  3. If redis-server is not already running, open an additional console window and run redis-server or redis-server & to start the Redis server in the background. The demos assume default login, no password.

##Running demo 01 -- Channel example

This demo shows how to send a series of URLs to a consumer process that computes an SHA1 value for each URL.

  1. In the first console window Run node worker01.js. It will wait for some data to appear in the queue.
  2. In the second console window, run node provider01.js, which will place four URLs in the queue. Shortly thereafter, the worker01 process will pick up the four URLs and display them, fetch a page body for each, and compute an SHA1 value for each.
  3. Repeat step 2 a few times.
  4. In the second console window, run node provider01.js stop, which will put a stop command in the queue. Shortly thereafter, the worker01 process will stop.

##Running demo 02 -- Channel example

This demo shows how to send a series of URLs to a consumer process that computes an SHA1 value for each URL and returns the SHA1 result to the provider process.

  1. In the first console window Run node worker02.js. It will wait for some data to appear in the queue.
  2. In the second console window, run node provider02.js 01, which will place four URLs in the queue. Shortly thereafter, the worker02 process will pick up the four URLs, display them, fetch a page body for each, and compute an SHA1 value for each, and then return the SHA1 result to the provider02 instance, which will display the result.
  3. Repeat step 2 a few times.
  4. In the second console window, run node provider02.js stop, which will put a stop command in the queue. Shortly thereafter, the worker02 process will stop.

##Running demo 03 -- WorkQueueMgr example

This demo shows how a worker can service multiple queues using WorkQueueMgr. The provider process, by default, sends three strings to one queue and three strings to the other.

  1. In the first console window Run node worker03.js. It will wait for some data to appear in the queue.
  2. In the second console window, run node provider03.js, which will place three strings in each queue. Shortly thereafter, the worker03 process will pick up the six strings from their respective queues and display them.
  3. Repeat step 2 a few times.
  4. In the second console window, run node provider03.js stop, which will put a stop command in the queue. Shortly thereafter, the worker03 process will stop.

Note that, when running worker03, one optionally may use a 'mem' parameter to monitor memory usage. For example:

node worker03.js mem | grep '>>>'

When monitoring memory usage, run node provider03.js 400 repeatedly, say as many as 50 times, to pump a sufficient amount of data to worker03 to detect any leaks. Sample memory usage output:

    >>>current = 3118200, max = 0
    >>>current = 3248152, max = 0
    >>>current = 3265896, max = 3265896
    >>>current = 3214184, max = 3265896
    >>>current = 3469112, max = 3469112
    >>>current = 3474064, max = 3474064
    >>>current = 3466856, max = 3474064
    >>>current = 3471904, max = 3474064
    >>>current = 3470080, max = 3474064

##Running demo 04 -- WorkQueueMgr example

This demo is almost the same as demo 02 but uses WorkQueueMgr rather than Channel. It shows how to send a series of URLs to a consumer process that computes an SHA1 value for each URL and returns the SHA1 result to the provider process.

  1. In the first console window Run node worker04.js. It will wait for some data to appear in the queue.
  2. In the second console window, run node provider04.js 01, which will place four URLs in the queue. Shortly thereafter, the worker04 process will pick up the four URLs, display them, fetch a page body for each, and compute an SHA1 value for each, and then return the SHA1 result to the provider04 instance, which will display the result.
  3. Repeat step 2 a few times.
  4. In the second console window, run node provider04.js stop, which will put a stop command in the queue. Shortly thereafter, the worker04 process will stop.

To check out the arity feature, try the above again using node worker04.js 3 in step 1. Observe that the worker will process three input requests in parallel and that the results may become available in a different order than the input requests.

To check out the timeout feature, try the above yet again using node worker04.js 1 1. Observe the timeout messages emitted about once every second.