Administrator Guide

Solution Architecture
16 Dell EMC Ready Solution for HPC PixStor Storage | Document ID
adapters, at least one of those adapters must be connected to the PixStor solution to get access to the file
system and any information it has stored (two connections if redundancy is required on a single gateway). In
addition, the gateways can be connected to other networks adding NICs supported by the PowerEdge R740
on the four x8 slots available (one x8 slot is used by a PERC adapter to manage local SSDs for the OS).
The Samba’s Clustered Trivial Data-Base (CTDB) is a clustered database used to manage the NFS and SMB
services on the gateway nodes, providing high availability, load balancing and monitoring of the nodes in the
CTDB cluster. For each of the gateways in the CTDB cluster, a Domain Name System (DNS) entry with an A
record for their IP is added, such that all have the same hostname, a sort of “public Gateway name.” That
Gateway name is then used by clients to mount those services, that way the name server daemon (named)
can assign all the gateways in the CTDB cluster to clients in a round robin fashion. When needed, NFS-
Ganesha (an open source, user space, NFS file server) can be used as an alternative to the regular NFS
server services, and it is also managed by the CTDB cluster.
Behind the Gateways, a PixStor system must be accessed and exported to the clients. For characterizing the
gateways in this work, a PixStor solution with high demand metadata and the capacity expansion modules
was used. That is:
• For Storing metadata, 2x PE R740 servers were connected to both controllers of a single ME4024 fully
populated with 960 GB SAS3 SSDs.
• For storing data 2x PE R740 servers were connected to all controllers of 4x DellEMC PowerVault (PV)
ME4084 disk arrays, each of them with an expansion box, for a total of 4x PV ME484. All arrays fully
populated with 12TB SAS3 NLS HDDs.
Ngenea Nodes
The hardware for Ngenea nodes is exactly the same as for the Gateway nodes, but has different software
installed and requires a different license. Since these nodes were not tested at the time of publishing this
work, a future blog will describe them in more detail and present some performance characterization and use
case relevant for a DellEMC Ready Solution for HPC PixStor Storage.
Advanced Analytics
Among PixStor capabilities, monitoring the file system via advanced analytics can be essential to greatly
simplify administration, helping to proactively or reactively find problems or potential issues. Next, we will
briefly review some of these capabilities.
Figure 7 shows useful information based on the file system capacity. The left side shows the file system total
space used, and the top ten users based on file system capacity used. The right side provides a historical
view with capacity used across many years, then the top ten file types used and top ten filesets, both based
on capacity used and in pareto chart diagram. With this information, it is easy to find users getting more than
their fair share of the file system, identify trends of capacity usage to decide future growth for capacity,
ascertain what files are using most of the space or what projects are taking most of the capacity.