IBM High Performance Computing Cluster Health Check - Important Points

To Download Link - DOWNLOAD

1)HPC cluster seems intimidating,many cluster problems are in fact easily resolved through careful verification steps.
2)cluster health check tools mostly test components and a little point-to-point network performance.
3)The purpose of a High Performance Computing (HPC) cluster is to solve scalable problems in a shorter time through parallelism.
4)The goal of verification stage(TESTING) is to gain confidence in hardware and software before introducing it to the user.
5)healthy cluster is built from the bottom to the top. Therefore, you first must make sure that each single device works
as expected before performing the next step.
This approach leads to the pyramid model of verification shown in Figure 3-2 on page 34.

IB 100% Non Blocking FAT TREE Topology Inter Connect Architecture Formula

1)IB Non Blocking Mode: Source Ports has the dedicated Channel to the destination ports without looping.

2)IB Non Blocking is made through FAT Tree Topology.

FAT Tree Topology Is one of the method.

Using 36 Port IB Switch
SS	LS	Max No Connection (or) Ports	Dedicate Connectivity Between SS & LS
SS	LS=SS* 2	LS*18	Possibility=36/LS=0
2	4	72	9
3	6	108	6
4	8	144	4.5
5	10	180	3.6
6	12	216	3
7	14	252	2.57
8	16	288	2.25
9	18	324	2
10	20	360	1.8
11	22	396	1.64
12	24	432	1.5
13	26	468	1.38
14	28	504	1.29
16	32	576	1.13
17	34	612	1.06
18	36	648	1
19	38	684	0.95
NOTE 1 : SS – Spine Switch LS – Leaf Switch NOTE 2 : Possible IB Switch connectivity are made in the light YELLOW color