Viewing Analysis Results

Prerequisites

An HPC cluster check task has been created and the analysis is complete.

Procedure

In the System Profiler area on the left, click the name of the target analysis task.
The node list is displayed.

Click the name of the target node to view the analysis results.

Click a node name. The Cluster Configuration tab page is displayed by default, as shown in Figure 1. Table 1 describes the parameters.

You can move the mouse pointer to next to a group to view the grouping basis or click to group data as required.
Click Filter Node Data to view the detailed data of a node or Clear Node Filter to clear the filter conditions.
Parameters collected by a privileged user are marked with asterisks (*).

Figure 1 Cluster hardware configuration (CPU)

**Table 1** Parameter description
Parameter	Description
CPU
Node IP Address	IP address of a node in the cluster.
Data Group	Group to which the current node belongs.
CPU Model	CPU model of the node.
Logical CPUs	Number of logical CPUs of the node.
Physical CPUs	Number of physical CPUs of the node.
Cores Per Physical CPU	Number of cores per CPU on the node.
Hyper-Threading	Number of hyperthreads on the node.
L1i cache	L1i cache.
L1d cache	L1d cache.
L2 cache	L2 cache.
L3 cache	L3 cache.
NUMA Nodes	Number of NUMA nodes.
Dominant Frequency	Dominant frequency of the CPU.
Cache Line Size	Cache line size of the CPU.
Vendor	CPU vendor.

Figure 2 Cluster hardware configuration (GPU)

**Table 2** Parameter description
Parameter	Description
Node IP Address	IP address of a node in the cluster.
Data Group	Group to which the current node belongs.
CUDA Version	CUDA version of the node.
Driver Version	GPU driver version of the node.
GPUs	Number of GPUs on the node.
Model	GPU model.
SM Clock Frequency	SM clock frequency of the GPU.
Memory Clock Frequency	Memory clock frequency of the GPU.

Figure 3 Cluster hardware configuration (memory)

**Table 3** Parameter description
Parameter	Description
Node IP Address	IP address of a node in the cluster.
Data Group	Group to which the current node belongs.
Total Memory Capacity	Total memory capacity of the node.
Page Size	Page size of the node.
Huge Page Size	Huge page size of the node.
*Vendor	Vendor of the DIMMs on the node. The data is collected by privileged users.
*Model	Model of the DIMMs on the node. The data is collected by privileged users.
*Size	Memory capacity of the DIMMs on the node. The data is collected by privileged users.
*Width	Bit width of the DIMMs on the node. The data is collected by privileged users.
*Clock Frequency	Clock frequency of the DIMMs on the node. The data is collected by privileged users.
*Bank Locator	Location information of the DIMMs on the node. The data is collected by privileged users.

Figure 4 Cluster hardware configuration (network)

**Table 4** Parameter description
Parameter	Description
Network Device
Node IP Address	IP address of a node in the cluster.
Data Group	Group to which the current node belongs.
RDMA Supported	Indicates whether Remote Direct Memory Access (RDMA) is supported on the node.
Logical NICs/Ports	Number of logical NICs or network ports on the node.
Logical Name	Logical name of the NIC on the node.
Model	Model of the NIC on the node.
Driver	NIC driver of the node.
Rate	NIC speed of the node.
MTU	Maximum size of data packets that can pass through the network port of the node.
Txqueuelen	Storage length of the buffer for data transmission on the network port of the node.
*PF_LOG_BAR_SIZE	Value of NIC parameter PF_LOG_BAR_SIZE. The data is collected by privileged users.
Driver Version	Driver version of the NIC on the node.
*Clock Frequency	Clock frequency of the NIC on the node. The data is collected by privileged users.
Device Name	Device name of the NIC on the node.
DSCP Priority	Differentiated Services Code Point (DSCP) value of the NIC on the node.
*TOS: service type	Service type of the NIC on the node. The data is collected by privileged users.
Priority Trust Mode	Priority trust status of the NIC on the node.
Flow control priority	Information about the priority-based flow control (PFC) of the NIC on the node.
DSCP	DSCP details of the NIC on the node.
DCQCN	DCQCN details of the NIC on the node.
Port	Port number of the network port on the node.
GID	GID of the network port on the node.
Version	Version of the network port on the node.
Index	Index of the network port on the node.
IP	IP address of the network port on the node.
Transfer Protocol	Transfer protocol of the network port on the node.
Device name: Port	Device and port number of the network port on the node.
Latency	Latency of the network port on the node.
Overhead	Overhead used by the network port on the node.
Bandwidth	Bandwidth of the network port on the node.
RX	Size of the message receiving queue of the NIC on the node.
TX	Size of the message transmit queue of the NIC on the node.
Other	Other configurations of the NIC on the node.
Combined	Number of enabled queues of the NIC on the node.
Route Configuration
Node IP Address	IP address of a node in the cluster.
Data Group	Group to which the current node belongs.
Gateway	Gateway of the current node.
Genmask	Mask of the current node.
Destination	Destination IP address of the route on the current node.
Flags	Flag of the route on the current node.
Metric	Route distance of the current node.
Ref	Number of references to routing entries of the current node.
Use: number of times that the routing software queries the router	Number of times that the routing entry of the current node is searched by routing software.
Iface	Output interface corresponding to the routing entry of the current node.
Hosts Configuration
Node IP Address	IP address of a node in the cluster.
Host IP	Host IP address corresponding to the node IP address.
Host Name	Host name corresponding to the node IP address.

Figure 5 Cluster hardware configuration (drive)

**Table 5** Parameter description
Parameter	Description
Node IP Address	IP address of a node in the cluster.
Data Group	Group to which the node belongs.
Total Drive Capacity	Total drive capacity of the node.
Drives	Number of drives on the node.
Name	Drive name.
Vendor	Drive vendor.
Capacity	Drive capacity.
Type	Drive type.
Model	Drive model.
System disk partition directory	System drive partition directory of the node.
Partition type	Partition type of the system drive on the node.
File system type	File system type in the system drive partition directory on the node.
Capacity	Target capacity of the system drive partition on the node.

Figure 6 Cluster hardware configuration (interconnection)

**Table 6** Parameter description
Parameter	Description
Node IP Address	IP address of a node in the cluster.
Data Group	Group to which the node belongs.
*Model	Model of the extended device on the current node. The data is collected by privileged users.
*Width	Bit width of the extended device on the current node. The data is collected by privileged users.
*Clock Frequency	Clock frequency of the extended device on the current node. The data is collected by privileged users.
*Capabilities	Capabilities of the extended device on the current node. The data is collected by privileged users.
*RAID Card Model	RAID card model of the extended device on the current node. The data is collected by privileged users.
*RAID Width	RAID card bit width of the extended device on the current node. The data is collected by privileged users.
*RAID Clock	RAID card clock frequency of the extended device on the current node. The data is collected by privileged users.
*Drive Name	Drive name of the extended device on the current node. The data is collected by privileged users.

Figure 7 Cluster software consistency distribution (OS)

**Table 7** Parameter description
Parameter	Description
NUMA
Node IP Address	IP address of a node in the cluster.
Data Group	Group to which the current node belongs.
Policy	NUMA policy of the current node.
Preferred Node	Preferred NUMA node of the current node.
Resource Restriction
Node IP Address	IP address of a node in the cluster.
Data Group	Group to which the current node belongs.
Core File Size	Size of the core file on the current node.
Data Seg Size	Maximum size of a data segment on the current node.
Scheduling Priority	Scheduling priority of the current node.
File Size	Maximum size of the file on the current node.
Pending Signals	Maximum number of pending signals on the current node.
Max. Locked Memory	Maximum locked memory of the current node.
Max. Memory Size	Maximum memory of the current node.
Open Files	Maximum number of files that can be opened on the current node.
Pipe Size	Maximum pipe size of the current node.
POSIX Message Queues: includes the extra overhead of message queues	POSIX message queue size of the current node, including the extra overhead of the message queue.
Real-time Priority	Real-time queue priority on the current node.
Stack Size	Maximum stack size of the current node.
CPU Time	CPU time limit of the current node.
Max. User Processes	Maximum number of user processes on the current node.
Virtual Memory	Maximum size of the virtual memory on the current node.
File Locks	Maximum number of file locks on the current node.
Kernel Config
Node IP Address	IP address of a node in the cluster.
Data Group	Group to which the current node belongs.
Parameter	Kernel configuration parameters of the current node.
Value	Values of kernel configuration parameters of the current node.

Figure 8 Cluster software consistency distribution (software package)

**Table 8** Parameter description
Parameter	Description
Node IP Address	IP address of a node in the cluster.
Data Group	Group to which the node belongs.
Kernel	Kernel version of the node.
OS	OS version of the node.
DPC Client	Number of DPC connections on the node.
Donau Scheduler Agent	Number of Donau Scheduler agent connections on the node.
Mellanox Driver	Mellanox driver version of the node.
PostgreSQL	PostgreSQL database version of the node.
*BIOS	BIOS version of the node. The data is collected by privileged users.
Haveged	Haveged version of the node.

Figure 9 Cluster software consistency distribution (environment variables)

Figure 10 Cluster software consistency distribution (dependency libraries)

Figure 11 Cluster software consistency distribution (module)

**Table 9** Parameter description
Parameter	Description
Environment Variable
Node IP Address	IP address of a node in the cluster.
Data Group	Group to which the current node belongs.
LD_LIBRARY_PATH	Value of the LD_LIBRARY_PATH environment variable of the node.
INCLUDE	Value of the INCLUDE environment variable of the node.
LOADED_MODULES	Value of the LOADED_MODULES environment variable of the node.
Dependency Library
Node IP Address	IP address of a node in the cluster.
Data Group	Group to which the current node belongs.
MPI Library	MPI library used by the node.
CUDA Library	CUDA library used by the node.
Module
Node IP Address	IP address of a node in the cluster.
Data Group	Group to which the current node belongs.
Version	Version of the module used on the node.
List	Details about the module used on the node.

Figure 12 Cluster software consistency distribution (MPI/OpenMP)

(Optional) Compare data by group.
You can click to customize comparison groups. A maximum of 10 groups can be selected.

Figure 13 Filtering groups for comparison
(Optional) Filter node data.
You can click Filter Node Data to view specific data or Clear Node Filter to view all data. Figure 14 shows an example.

There are four filter options:

Include all: nodes and data matching all filter items

Include any: nodes and data matching one or more filter items

Exclude all: excluding nodes and data matching all filter items

Exclude any: excluding nodes and data matching any of the filter items

Figure 14 Data filtering

Parent topic: HPC Cluster Check