Today we continue our discussion of the  paper titled "Pelican: a  building block for exascale cold data storage". Pelican treats a group of disks as a single schedulable unit. Resource restrictions such as power consumption, vibrations and failure domains are expressed as constraints over these units.

With the help of resource constraints and scheduling units, Pelican aims to be better than its over provisioned counterpart racks using computations in software stacks

Pelican uses resources from a set of resource domains which is a subset of disks. Pelican proposes a data layout and IO scheduling algorithms by expressing these resource domains as constraints over the disks. 

Pelican uses a metadata service called Catalog that is durable and highly available. It maintains bookkeeping information such as the mapping from a blob key to the 18 disks and group which store metadata. It is updated during write, rebuild and delete requests and looked up during read requests.

#codingexercise

Write  a method to delete all the nodes from a binary tree that lie on a path whose sum from root to leaf is less than a given value K.  

Void delPathSums(Node root, int K, ref int cur) 

{ 

If (Root == null) return; 

Cur+= root.data; 

Int left = cur; 

Int right = cur; 

Root.left = delPathSums(root.left , K, ref int left); 

Root.right = delPathSums(root.right, K, ref int right); 

Cur += max(left, right); 

If (cur <k){ 

Tree_delete(root); 

Root = null; 

} 

Return root; 

}

#DataDaySeattle2016 continued

Hi,

Here's the link to the file:

https://1drv.ms/w/s!Ashlm-Nw-wnWk3yjDmBBSqrd4PSr

Shared from Word for Android

https://1drv.ms/w/s!Ashlm-Nw-wnWk3y1Lm87oOphR2j6