[ 
https://issues.apache.org/jira/browse/HIVE-23085?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17067877#comment-17067877
 ] 

Gopal Vijayaraghavan commented on HIVE-23085:
---------------------------------------------

Here's the actual implementation known to work instead of using multiple 
independent disks (on-prem the independent disks are marginally better, since 
we can keep limping using partial hardware - on-cloud, it is better to give up 
and get a better box ASAP).

{code}
            # mount nvme SSDs as LVM to /data when available for specific 
instance types
            # There are max 4 nvme slots in r5 instances. /dev/nvme0n1 is root 
EBS volume.
            # concat all existing nvme slots to logical volume creation
            # 
https://stackoverflow.com/questions/44458304/how-to-escape-in-cloudformations-fnsub
            if [ -e /dev/nvme1n1 ]
            then
              for i in $(seq 1 4); do
                if [ -e /dev/nvme${!i}n1 ]
                then
                  x+=" /dev/nvme${!i}n1"
                fi
              done
              yes | sudo pvcreate $x
              yes | sudo vgcreate nvmevg $x
              yes | sudo lvcreate -l 100%FREE nvmevg -n nvmelv
              yes | sudo mkfs.ext4 /dev/nvmevg/nvmelv
              sudo mkdir -p /data
              sudo mount /dev/nvmevg/nvmelv /data
            fi
{code}

> LLAP: Support Multiple NVMe-SSD disk Locations While Using SSD Cache
> --------------------------------------------------------------------
>
>                 Key: HIVE-23085
>                 URL: https://issues.apache.org/jira/browse/HIVE-23085
>             Project: Hive
>          Issue Type: Improvement
>            Reporter: Syed Shameerur Rahman
>            Assignee: Syed Shameerur Rahman
>            Priority: Major
>              Labels: pull-request-available
>             Fix For: 4.0.0
>
>         Attachments: HIVE-23085.01.patch
>
>          Time Spent: 10m
>  Remaining Estimate: 0h
>
> Currently we can configure only one SSD location while using SSD cache in 
> LLAP. This highly undermines the capacity of some machines to use its disk 
> capacity to the fullest. For example *AWS* provides *r5d.4x large* series 
> which comes with *2 * 300 GB NVme SSD disk* with the current design only one 
> of the mounted *NVme SSD* disk can be used for caching. Hence adding support 
> for caching data at multiple ssd mounted locations.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

Reply via email to