CES/NFS/SMB Function Enhancements

1) CES FO/FB timeout, shorten the io pending time in the failure of CES node failure/GPFS node failure/one of FG failure in replication mode. The io pending time should be less than 4-5s. 

2) AFM/DR RPO Interval, the minimal RPO interval is 60 minutes, customer expect to set it to 1-5 minutes. 60 minutes RPO interval mean customer maybe loss 60 minutes data during failover to secondary site. 1-5 minutes RPO interval is acceptable in most of scenarios.

3)Decrease memory requirement in CES.The Scale FAQ list CES node need 128GB memory. It's huge to most of SMB scenarios. Most of  Customer often have 50/100/200 SMB/NFS per CES ndoe pair. Customer hope to support  16G/32G/64G/128G memory models in CES node. If it's no, customer want to know how many  memory size that can support 50/100/200 sessions(NFS/SMB/NFS+SMB) ? 

4)CES interface bonding,the current design is alias the CES IP to one of ethernet interface of alias to one of ethernet interface with fixed IP. This will waste of static IP address in large environment. Customer hope to support CES IP assign to specific ethernet interface.

5)NFS/SMB transparent failover, CIFS/NFS session reconnect automatically during CES FO/FB.

6)Snapshot numbers, Today, we only support 256 Global snapshots, either filesystem level or fileset level. Customer hope to support 2048 or 4096 snapshots. 

7) Schedule snapshot, support schedule snapshot in 15s or 1 minutes interval.

8)Snapshot target support R/W mode. Today, snapshot target is read only mode. Customer hope to support read/write in snapshot target directory.