Shulou(Shulou.com)06/02 Report--

We often say that "there is no end to performance optimization", and the ultimate user experience has become the golden rule of IT application design and development. Although there is no upper limit to the user experience, its bottom line has always been clear. As Luo Wei, general manager of Dynatrace Greater China, said at Dynatrace Perform 2018, "the most perfect user experience that can be achieved at present is actually the bottom line of the next experience." All companies seem to be busy with digital transformation, which makes manufacturers uneasy. The emergence of a large number of cloud applications, containers and micro-services makes traditional APM defenseless, and these manufacturers are looking for transformation. How to keep the bottom line of this user experience? How to raise this bottom line step by step? What is the right way to open cloud computing, AI, and even augmented reality in the field of performance monitoring?

Distinguished guests for this issue:

Chief Product Officer of Andreas Lehofer,Dynatrace

Rafi Katanasho, Chief Technology Officer of Dynatrace Asia Pacific and Vice President of solution sales.

Traditional APM is at a dead end, and cloud monitoring games emerge one after another!

If you give yourself only 5 square meters of space, you may hit a brick wall everywhere; if you give yourself a whole piece of land, your field of vision will be broadened. As an elder in the field of APM, Dynatrace used to keep an eye on the so-called $3 billion APM market. After the baptism of digital transformation, Dynatrace realized that traditional APM is only suitable for a world that no longer exists, and cloud computing and artificial intelligence are the real battlefields of APM.

In retrospect, cloud computing actually gives priority to digital transformation. Cloud computing has not only changed the technology stack of enterprises, but also changed the environment of IT operation and maintenance. Micro-services and container technologies have increased, application complexity has increased, iteration cycle has been greatly shortened, and there are a variety of cloud and deployment methods within enterprises. Although the entire IT environment of the enterprise does not have to be converted to the cloud model, it is part of the transformation, and we need to understand the state of the entire cloud environment and the transformation needs. Not only enterprises are at a loss, but manufacturers are also facing the hurdle of transformation. After three product iterations before and after Dynatrace, the emergence of the third generation product OneAgent is really in line with the concept of cloud monitoring.

OneAgent is easy to deploy, and can automatically discover the dependencies among all host services throughout the platform, including the related services in the container. Users can view the corresponding management areas according to different permissions. Once there is a problem, the traditional alarm method will alert the user of a single event, while Dynatrace can view the whole process of user access, and display the impact scope of this event, the number of users and other information. Of course, the whole process fully respects the privacy of users, and users have the right to choose the range of data available to the machine.

Rafi Katanasho, Chief Technical Officer of Dynatrace Asia Pacific and Vice President of solution sales

Although everyone has a different understanding of AI, it is true that AI has been abused. Rafi believes that the AI algorithm is not a simple way to get and analyze the data, especially in the field of performance monitoring. Dynatrace's AI algorithm has higher requirements for data quality and value, and has clear requirements for data quality in the process of data acquisition, and will establish the relationship between data according to causality, use specific algorithms to calculate in the field of performance management, and improve data value through continuous optimization of the algorithm.

In addition to AI,Dynatrace 's attempt to combine performance monitoring with HoloLens, Andreas says this allows business people who don't understand technology to get a visual picture of the performance of current applications through simple charts or aggregate data. Of course, all of this requires access to user data, and privacy has become a major point that has to be discussed.

Performance monitoring cannot bypass "fetching data", so how can user privacy be guaranteed?

When it comes to performance monitoring and artificial intelligence, data is an unavoidable topic. By obtaining user experience data, Dynatrace can understand every action and possible trigger result of the user. As the saying goes, Dynatrace only acquires data related to the user's performance experience, and users have the right to choose which part of the data can be obtained and which part of the data can not be touched. As we all know, the requirements for data privacy in Europe are very strict, and most of the development teams of Dynatrace are in Europe, so they attach great importance to data security and user privacy.

Andreas Lehofer, Chief Product Officer of Dynatrace

Among the many customers of Dynatrace, SAP is the largest and most representative. From a technical point of view, no one can deny the strength of SAP, but they give up self-built monitoring system to use Dynatrace, on the one hand, because of cost considerations, on the other hand, SAP needs more than just a monitoring platform. Andreas revealed that SAP hopes to turn monitoring into a part of the overall business strategy, automatically monitor more than 20, 000 applications on its platform and follow the principle of API priority, and comb the data of the entire platform through AI engine. This is the reason why SAP gives up self-built monitoring system, which is not just a problem of performance monitoring, but the result of software intelligence. This is the end result of Dynatrace's transformation-becoming a software intelligent enterprise.

Conclusion

When it comes to software intelligence, Andreas said that if Dynatrace focuses on APM, it will not be scalable in the future. In the future, Dynatrace hopes to become a software intelligent enterprise, which can help customers get better software experience and conduct intelligent analysis, and conduct more in-depth research at the operation and maintenance and DevOps level. The traditional $3 billion APM market no longer exists, but the new and larger cloud market has more room for growth, which is the direction of Dynatrace.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.