Free online consultation available. Please feel free to contact us.

deadlock

  • Glossary

"deadlock"

This glossary explains various keywords that will help you understand the mindset necessary for data utilization and successful DX.
This time, we will explain "deadlock," a basic knowledge in computer science regarding the behavior of processing on a computer.

What is deadlock?

A deadlock is a problem in which multiple processes (threads, processes, etc.) that use multiple resources cannot proceed. Each process is waiting for the release of a resource that the other process is using exclusively, and no process can proceed.
When using your computer, you may experience applications becoming unresponsive and "freezing," which may be due to a deadlock.

How deadlocks occur

Deadlock is a term often used in computer science, but it is a problem that generally occurs in any situation where something is being done, not just in computers. Here, we will introduce the situations in which this problem occurs using examples from everyday tasks.

Let's say you have a task called "cut vegetables" that involves the following steps:

  • Secure a cutting board
  • Secure the knife
  • Cutting vegetables
  • Put the knife back
  • Put the cutting board back

This is not a particularly bad workflow, but if "Person A" and "Person B" try to cut vegetables at the same time, this is what might happen.

Mr. A

  • Procedure: Secure a cutting board
    ⇒ Person A secures the cutting board
  • Procedure: Secure a knife
    ⇒Mr. B reserved it first, so it is in use and cannot be used.

Waiting for the knife to become available

Mr. B

  • Procedure: Secure a cutting board
    ⇒The cutting board cannot be reserved because it is currently being used by A.
  • Procedure: Secure a knife
    ⇒Mr. B secures the knife

Waiting for a cutting board to become available

If this happens, both A and B will be waiting forever. This state is called a deadlock. In relation to the explanation above, "A" and "B" correspond to "processes" such as processes and threads, while the "knife" and "cutting board" correspond to "resources" that are used exclusively.

Humans A and B would not be foolish enough to wait in this state for a long time, but since computer programs only do what they are instructed to do, when they get to this state they will freeze and not be able to proceed with the processing.

Deadlock prevention

There are several ways to prevent deadlocks. Below are some of the most common ones.

  • Determine the order in which resources are acquired:
    In the example above, there is a way to ensure that resources are secured in that order by ordering the resources so that the "cutting board" is secured first and then the "knife."
  • Prioritize your actions:
    For example, one way to do this is to rank processes by priority, such as giving person A a higher priority than person B, and then create a rule that person A will be given priority in using resources even if it means taking resources away from person B.
  • Bundle resources:
    The deadlock problem arises because "cutting boards" and "knives" are separate resources, so one way to solve this is to combine "cutting boards and knives" into one resource.
  • Release resources after a certain time:
    If the waiting state persists after waiting for a while, you can set a rule to release the resources you have secured.

However, it is not easy to prioritize all resources, and it is possible that someone with a low priority may never be able to use the resource. The rules for releasing resources can also cause another problem (livelock) where multiple processes repeatedly secure and release resources indefinitely, preventing the process from progressing. In other words, the solution can sometimes create another problem.

Efforts to consolidate resources, for example, by making all of a company's equipment one resource, will prevent deadlocks from occurring, but this can also be difficult, as it means that the company will only be able to work one person at a time, making it inefficient.

Unfortunately, there is no simple and universal solution to deadlocks, so it is not possible to prevent deadlocks by simply creating an IT system in the normal way.

Deadlocks can be difficult to deal with and can be a serious problem

It is difficult to find "possible deadlocks"

Unfortunately, deadlocks can be difficult to detect as a problem. In the example above, the problem only occurs when Person A and Person B happen to want to cut vegetables at the same time. Because this is a problem that may or may not occur depending on the timing of the processing, even if an IT system is designed in a way that can cause a deadlock, it can be difficult to detect, and even if a problem does occur, it can be difficult to identify the cause or reproduce the problem.

Problems can occur at awkward times

Because deadlocks depend on the timing of processing, there are chance factors that can cause them to occur, and although a system may appear to be working without any problems, problems may only be discovered long after it has been running. Because conflicts in processing timing are more likely to occur during times of high load, the system may stop responding at the very time when it would be a problem if it froze up, such as during the end-of-year period when business processes are concentrated.

A serious problem for IT systems that are responsible for the safety, security and reliability of society

If your PC freezes while you're using it, you might think it can't be helped and just force shut it down and restart it, but if your IT system stops responding, it could become a major problem.

It would be a major incident if a bank's system stopped responding and bank transfers stopped, and it would result in an accident if the program that controls the flight of an airplane stopped responding. Problems like deadlocks must be prevented from occurring in IT systems where problems should not occur.

The need to "build a proper IT system" and the need for "in-house development"

In-house development efforts are becoming more popular these days. In a world of increasing change and uncertainty, where business is required to operate at a faster pace than ever before, it will become increasingly necessary for business sites to master the use of IT themselves. It is no longer uncommon for organizations to use the cloud on their own.

However, when we create IT for business purposes ourselves, we also have to deal with problems such as deadlocks. Unfortunately, IT systems that are created without much thought, or that are created by people who are not engineers, can sometimes cause problems.

  • Implement it properly and in-house to avoid problems such as deadlocks
  • Outsource to a technically capable external company (give up on in-house production)
  • In-house development utilizing infrastructure that is in place to prevent problems such as deadlocks

If you are an IT engineer by profession, you should naturally understand deadlocks, implement them in a way that prevents them from occurring, and pay attention to whether existing implementations have deadlock issues.In other words, you should understand them properly and create them properly.

However, things get complicated when it comes to in-house development. In many cases, it is not realistic for people on the front lines of business to acquire the skills to the point where they can say, "Our team members can implement something that doesn't cause deadlocks, so it's fine." In other words, it becomes difficult for them to "build it properly" by themselves.

However, outsourcing to external companies with the technical capabilities would mean giving up on in-house production, and we would be back to the same situation as before.

In reality, I think it would be better to develop a "well-built IT system" as a foundation and use it to develop in-house. If the foundation itself has been designed to address and consider issues that require attention, such as deadlocks, then by using the foundation for in-house development, you can create an IT system that has been designed with these issues and considerations in mind.

Is that no-code/low-code product okay?

In in-house development efforts rarely involve full-scale in-house programming, and I think that what is often used as the means (or foundation) for in-house development is so-called no-code or low-code products, such as the use of some kind of SaaS (such as kintone or Salesforce).

Some of these products are "easy to use but can only do simple things."Even if they are easy to create, they cannot be fully implemented and you cannot create what you need to create, or they cannot process as the amount of data increases due to insufficient processing power, they do not operate stably due to processing freezing due to deadlocks, and if they have security issues, they cannot be said to be IT that can be trusted to handle your business.

Unfortunately, some no-code products and services may look modern and state-of-the-art, but they are little more than hobbyist programming.

If you are going to go in-house, you need to base your implementation on products suitable for professional use, with a proven track record and functionality that are safe, reliable, and can be trusted to handle the job properly.

Are you interested in "iPaaS" and "connecting" technologies?

The key to successfully implementing "cloud utilization," "in-house data utilization," and "business automation" is to establish a "data integration platform" that can be developed in-house by your company.

Try out our products that allow you to freely connect various data and systems, from on-premise IT systems to cloud services, and make successful use of IT.

The ultimate "connecting" tool: data integration software "DataSpider" and data integration platform "HULFT Square"

"DataSpider," data integration tool developed and sold by our company, is a "connecting" tool with a long history of success as the foundation supporting various companies' business systems. "HULFT Square," a data integration platform, is a "connecting" cloud service developed using DataSpider technology.

Another feature is that development can be done using only the GUI (no code), without writing code like in regular programming, so business staff who have a good understanding of their company's business can take the initiative to utilize it.

Moreover, it can be used using only a GUI, and is easy enough for even non-programmers to use, while still providing "high development productivity" and "full-fledged performance that can serve as the foundation for business (professional use)."

It can smoothly solve the problem of "connecting disparate systems and data" that hinders successful IT utilization. We offer a free trial version and online seminars where you can try it out for free, so we hope you will give it a try.

Try outDataSpider/ HULFT Square 's "connecting" technology:

Glossary Column List

Alphanumeric characters and symbols

A row

Ka row

Sa row

Ta row

Na row

Ha row

Ma row

Ya row

Ra row

Wa row

»Data Utilization Column List

Recommended Content

Related Content

Return to column list