Hewlett-Packard (HP) created and markets the server-based operating system known as HP-UX. It is a Unix-based operating system built for Hewlett-Packard 9000 series servers. In corporate data centers, HP-UX is a popular operating system because it is reliable, secure, and stable. Mission-critical applications such as data warehousing, customer relationship management (CRM), and enterprise resource planning (ERP) systems are well-suited for this. HP-UX is not only used for High-Performance Computing (HPC) Clusters but is also used for web and application servers. It includes advanced features such as system partitioning, high availability clustering, and secure networking, and is optimized for HP server hardware. For mission-critical applications, HP-UX is a reliable and secure platform, and its scalability allows it to meet the requirements of organizations of all sizes.
Performance counters for the HP-UX operating system
System
- CPU User: CPU time used for routine and low-priority tasks. % breakdown of CPU time used.
- System CPU: The amount of time used to execute the kernel code. time spent on CPU as a percentage.
- CPU idle: CPU idle. % breakdown of CPU time used.
- Interruptions Count : Device interrupts per second, measured in interrupts (no clock).
- Context Switch Rate: The frequency of context transitions.
- Number of context switches: The number of context switches that each CPU performs per second.
- Processes that can be executed: the number of processes that are waiting to be executed.
- Processes that can be executed by CPU: Number of processes that each CPU has available to execute.
- Processes that sleep: number of processes that rest well.
- Processes Swapped: The total number of processes running or idle (less than 20 seconds).
Memory
- Active virtual pages: Virtual pages that are currently active. If a virtual page is open, it is said to be active.
- Free List Size: Size of the available list, or the free list.
- pages retrieved every second: Pages retrieved every second should be reported as page fault information. (every 5 seconds on average).
- Correct translation errors: Provide details of any page errors. Fix translation errors per second. (every 5 seconds on average).
- Processes swapped in: Processes Exchanged: processes exchanged per second. (every 5 seconds on average).
- Processes Exchanged: processes exchanged per second. (every 5 seconds on average).
- Pages received per second: Pages received. (every 5 seconds on average).
- Pages per second: pages without pages. (every 5 seconds on average).
- Pages Freed Per Second: Pages freed per second. (every 5 seconds on average).
- Expected short-term memory loss: Expect short-term memory loss. (every 5 seconds on average).
- Scan pages using a clock algorithm. The pages per second are scanned by the clock algorithm. (every 5 seconds on average).
Disk (per device)
- Kilobytes Sent and Received Per Second: The number of kilobytes sent and received per second.
- Searches per second: The frequency of the searches.
- Average Seek Time in Milliseconds: The average seek time in milliseconds.
Process
- The portion of the task’s CPU time that has elapsed since the last update, expressed as a percentage of all CPU time is called Process Process CPU Time.
- Process memory usage: (KB) resident size. the physical memory that a job has not swapped.
Network (per interface)
- Incoming bytes/s: The amount of data that the network interface receives per second in terms of incoming bytes/s.
- Incoming packets/s: The number of packets that the network interface receives each second is known as incoming packets.
Tcp
- Incoming segments/s: The total number of TCP segments received per second is known as Incoming TCP Segments/s.
- Segments completely duplicated/s : Fully Duplicate Segments: The number of duplicate segments received each second. When a segment is received many times after one or more retransmissions, it is called a duplicate.
- % Segments duplicated: The segments that are fully duplicated divided by the total number of segments received is the percentage of duplicated segments.
- Outgoing segments/s: The number of TCP segments sent each second is known as the outgoing segments.
- Segments retransmitted/s: Number of TCP segments retransmitted every second, measured in segments/s retransmitted.
- % Segments retransmitted: Rebroadcast Segments / Outgoing Segments *100 = % Rebroadcast Segments, which is the percentage of rebroadcast segments.
- Retransmit timeouts: Forwarding timeouts: how many timeouts do a segment broadcast?
