A method for real-time monitoring and automatic processing of online tasks

Abstract

The application provides a real-time monitoring and automatic processing method of online tasks, searches all Log generated under a folder within one hour in past running time; opens all searched Log one by one and searches a keyword "Successfully completed" line by line; opens again the Log searched with the keyword and scans and reads running time Runtime data line by line; and extracts task information of the Log with running time Runtime less than 20 seconds; processes numerical values needed in busb instruction; judges the type of test program Recipe; if the type of test program Recipe of the task with abnormal completion is OPC.DRC, only the task needs to be resubmitted, and the bkill instruction does not need to be executed; if the type of test program Recipe of the task with abnormal completion is TDOPC.DRC or OPCRP.DRC, the corresponding OPC.DRC or OPCV.DRC on line needs to be cleared first, and then two tasks are resubmitted; if the two tasks are submitted, the IDs of the two tasks need to be extracted from the queue, and the bmod instruction is used to establish the dependent relationship of the two tasks to determine the running order of the two tasks.

Description

Technical Field

[0001] This invention relates to the field of semiconductor technology, and in particular to a method for real-time monitoring and automatic processing of online tasks. Background Technology

[0002] In current tape-out work, we've observed that some tasks complete abnormally during execution; these abnormally completed tasks are termed "Done-abnormally." While the queue shows these tasks as finished, the execution logs show no required output. These tasks typically have a runtime of less than 20 seconds. This causes related tasks to remain in standby online, preventing engineers from seeing the necessary results the following day.

[0003] By reviewing the logs of the abnormal task, it can be inferred that the problem is most likely related to LSF (Low-Speed ​​File System), requiring contact with the vendor for resolution. However, since the company is no longer purchasing LSF, no vendor is providing maintenance services. Furthermore, the abnormal completion phenomenon occurs intermittently across different products, requiring engineers to manually rerun the process. If this occurs frequently, especially at night or during peak tape-out periods, it will severely impact the product's tape-out cycle. With the aim of minimizing additional expenses and engineer workload, we hope to use a programming language to replace engineers in promptly identifying and resolving the problem, minimizing its impact with minimal resources. Summary of the Invention

[0004] In view of the shortcomings of the prior art described above, the purpose of this invention is to provide a method for real-time monitoring and automatic processing of online tasks, which solves the problem in the prior art where tasks appear to have been completed from the queue, but no required data is found to have been output from the execution log, thus preventing engineers from seeing the required results.

[0005] To achieve the above and other related objectives, the present invention provides a method for real-time monitoring and automatic processing of online tasks, comprising at least:

[0006] Step 1: Search for all logs generated in the folder within the past one hour of runtime;

[0007] Step 2: Open each of the searched logs one by one and search for the keyword "Successfully completed" line by line;

[0008] Step 3: Reopen the Log containing the searched keywords and scan and read the Runtime data line by line; extract the task information from Logs with a Runtime of less than 20 seconds.

[0009] Step 4: Process the values ​​required by the busb command;

[0010] Step 5: Determine the type of the test program Recipe; if the type of the test program Recipe for the abnormally completed task is OPC.DRC, then you only need to resubmit the task and do not need to execute the bkill command; if the type of the test program Recipe for the abnormally completed task is TDOPC.DRC or OPCRP.DRC, you need to clear the corresponding OPC.DRC or OPCV.DRC online first, and then resubmit both tasks.

[0011] Step 6: If two tasks are submitted, the IDs of the two tasks need to be retrieved from the queue, and the dependency relationship between the two tasks needs to be established using the bmod command to determine the execution order of the two tasks.

[0012] Preferably, the values ​​required in the busb command in step four include: Calibre Version, Turbo, and Path.

[0013] Preferably, the values ​​required in the busb instructions in step four are captured from the Log where the runtime is less than 20 seconds. These values ​​are sorted and edited into the busb instructions required to submit the task.

[0014] Preferably, before executing step one, the program is set to run once every hour using the Linux Crontab command. When the program runs, step one is executed.

[0015] Preferably, in step three, the Log entry for the keyword "Successfully completed" indicates that the corresponding task has been completed online.

[0016] Preferably, in step three, a log entry with a runtime of less than 20 seconds is considered an abnormal completion.

[0017] As described above, the real-time monitoring and automatic processing method for online tasks of the present invention has the following beneficial effects: The present invention replaces engineers in real-time monitoring and processing of task execution, avoiding untimely processing and prolonged standby after abnormal task completion. It improves tape-out efficiency, ensuring tasks are effectively completed within the estimated time. During the trial period of this program, abnormally completed tasks can be resubmitted online and successfully completed, avoiding resource waste caused by needing to purchase additional services to solve the fundamental problem. More tasks can be adapted by adding test program types. Attached Figure Description

[0018] Figure 1 The diagram shown is a flowchart of the online task real-time monitoring and automatic processing method of the present invention. Detailed Implementation

[0019] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention.

[0020] Please see Figure 1 It should be noted that the illustrations provided in this embodiment are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0021] This invention provides a method for real-time monitoring and automatic processing of online tasks, comprising at least:

[0022] The main principle of online job real-time monitoring and automatic processing technology is to periodically monitor the completion status of jobs. When an abnormally completed job is detected, the system will automatically kill the associated job that has already been pent online and re-release the job online. Further, in this embodiment, before executing step one, the Linux Crontab command is used to set the program to run once every hour. When the program runs, step one is executed.

[0023] Step 1: Search the folder for all logs (run records of each task) generated within the past hour.

[0024] Step 2: Open each of the searched logs one by one and search for the keyword "Successfully completed" line by line;

[0025] Use the Linux crontab command to schedule the program to run once an hour. When the program runs, it first searches for all logs generated in the folder within the past hour. Open each log and search line by line for the keyword "Successfully completed." Finding this keyword indicates that the corresponding job has been completed online. Extracting the filenames from the logs that match the keyword reveals all the jobs completed in the past hour.

[0026] Step 3: Reopen the Log containing the searched keywords and scan and read the Runtime data line by line; extract the task information from Logs with a Runtime of less than 20 seconds.

[0027] Furthermore, in step three of this embodiment, a Log entry containing the keyword "Successfully completed" indicates that the corresponding task has been completed online. Furthermore, in step three of this embodiment, Log entries with a runtime of less than 20 seconds indicate abnormal completion.

[0028] It's necessary to determine if any of these jobs completed abnormally. By reviewing the logs of previously abnormally completed jobs, it can be found that these jobs typically had a runtime of less than 20 seconds. Therefore, during program execution, the logs will be reopened and scanned line by line to read the runtime data, and the information of jobs that completed within 20 seconds will be extracted for further processing.

[0029] Step 4: Process the values ​​required by the busb command;

[0030] Furthermore, in step four of this embodiment, the values ​​required by the busb command include: Calibre Version, Turbo, and Path.

[0031] Furthermore, in step four of this embodiment, the values ​​required for the busb instructions are retrieved from the Log where the runtime is less than 20 seconds. These values ​​are sorted and edited into the busb instructions required for submitting the task.

[0032] The system processes the values ​​required in the busb commands, including Calibre Version, Turbo, Path, CpuNumber, and Recipe Step. These values ​​can be retrieved from the log. After sorting and processing, these values ​​can be edited into the busb commands needed for the final job submission.

[0033] Step 5: Determine the type of the test program Recipe; if the type of the test program Recipe for the abnormally completed task is OPC.DRC, then you only need to resubmit the task and do not need to execute the bkill command; if the type of the test program Recipe for the abnormally completed task is TDOPC.DRC or OPCRP.DRC, you need to clear the corresponding OPC.DRC or OPCV.DRC online first, and then resubmit both tasks.

[0034] Step 6: If two tasks are submitted, the IDs of the two tasks need to be retrieved from the queue, and the dependency relationship between the two tasks needs to be established using the bmod command to determine the execution order of the two tasks.

[0035] If two jobs are submitted, the IDs of the two jobs need to be retrieved from the queue, and the bmod command is used to establish the dependency relationship between the two jobs to ensure the order in which the jobs are executed.

[0036] In summary, this invention replaces engineers in real-time monitoring and processing of task execution, avoiding delays and prolonged standby times after tasks complete abnormally. It improves tape-out efficiency and ensures tasks are effectively completed within the estimated timeframe. During the trial period, abnormally completed tasks could be resubmitted online and successfully completed, avoiding resource waste caused by procuring additional services to resolve the root cause. More tasks can be accommodated by adding test program types. Therefore, this invention effectively overcomes the various shortcomings of existing technologies and has high industrial applicability.

[0037] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A method for real-time monitoring and automatic processing of online tasks, characterized in that, At least including: Step 1: Search for all logs generated in the folder within the past one hour of runtime; Step 2: Open each of the searched logs one by one and search for the keyword "Successfully completed" line by line; Step 3: Reopen the Log containing the searched keywords and scan and read the Runtime data line by line; extract the task information from Logs with a Runtime of less than 20 seconds. Step 4: Process the values ​​required by the busb command; Step 5: Determine the type of the test program Recipe; if the type of the test program Recipe for the abnormally completed task is OPC.DRC, then you only need to resubmit the task and do not need to execute the bkill command; if the type of the test program Recipe for the abnormally completed task is TDOPC.DRC or OPCRP.DRC, you need to clear the corresponding OPC.DRC or OPCV.DRC online first, and then resubmit both tasks. Step 6: If two tasks are submitted, the IDs of the two tasks need to be retrieved from the queue, and the dependency relationship between the two tasks needs to be established using the bmod command to determine the execution order of the two tasks.

2. The method for real-time monitoring and automatic processing of online tasks according to claim 1, characterized in that: The values ​​required for the busb command in step four include: Calibre Version, Turbo, and Path.

3. The method for real-time monitoring and automatic processing of online tasks according to claim 2, characterized in that: In step four, the values ​​required by the busb instructions are retrieved from the logs where the runtime is less than 20 seconds. These values ​​are sorted and edited into the busb instructions required to submit the task.

4. The method for real-time monitoring and automatic processing of online tasks according to claim 1, characterized in that: Before executing step one, use the Linux Crontab command to set the program to run once every hour. When the program is running, execute step one.

5. The method for real-time monitoring and automatic processing of online tasks according to claim 1, characterized in that: The Log entry for the keyword "Successfully completed" found in step three indicates that the corresponding task has been completed online.

6. The method for real-time monitoring and automatic processing of online tasks according to claim 1, characterized in that: In step three, if the runtime is less than 20 seconds, the log indicates abnormal completion.

Images