[jira] [Commented] (TEZ-2496) Consider scheduling tasks in ShuffleVertexManager based on the partition sizes from the source

["Siddharth Seth (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/TEZ-2496?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14563701#comment-14563701 ] 

Siddharth Seth commented on TEZ-2496:
-------------------------------------

+1 for providing per partition information. This can also be used to re-distribute partitions to tasks so that all of them get the same amount of data. Providing this information to consumer tasks is also helpful - in terms of sizing buffers, and potentially prioritizing fetches.

> Consider scheduling tasks in ShuffleVertexManager based on the partition sizes from the source
> ----------------------------------------------------------------------------------------------
>
>                 Key: TEZ-2496
>                 URL: https://issues.apache.org/jira/browse/TEZ-2496
>             Project: Apache Tez
>          Issue Type: Improvement
>            Reporter: Rajesh Balamohan
>         Attachments: TEZ-2496.1.patch
>
>
> Consider scheduling tasks in ShuffleVertexManager based on the partition sizes from the source.  This would be helpful in scenarios, where there is limited resources (or concurrent jobs running or multiple waves) with dataskew and the task which gets large amount of data gets sceheduled much later.
> e.g Consider the following hive query running in a queue with limited capacity (42 slots in total) @ 200 GB scale
> {noformat}
> CREATE TEMPORARY TABLE sampleData AS
>   SELECT CASE
>            WHEN ss_sold_time_sk IS NULL THEN 70429
>            ELSE ss_sold_time_sk
>        END AS ss_sold_time_sk,
>        ss_item_sk,
>        ss_customer_sk,
>        ss_cdemo_sk,
>        ss_hdemo_sk,
>        ss_addr_sk,
>        ss_store_sk,
>        ss_promo_sk,
>        ss_ticket_number,
>        ss_quantity,
>        ss_wholesale_cost,
>        ss_list_price,
>        ss_sales_price,
>        ss_ext_discount_amt,
>        ss_ext_sales_price,
>        ss_ext_wholesale_cost,
>        ss_ext_list_price,
>        ss_ext_tax,
>        ss_coupon_amt,
>        ss_net_paid,
>        ss_net_paid_inc_tax,
>        ss_net_profit,
>        ss_sold_date_sk
>   FROM store_sales distribute by ss_sold_time_sk;
> {noformat}
> This generated 39 maps and 134 reduce slots (3 reduce waves). When lots of nulls are there for ss_sold_time_sk, it would tend to have data skew towards 70429.  If the reducer which gets this data gets scheduled much earlier (i.e in first wave itself), entire job would finish fast.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)