Shulou(Shulou.com)06/01 Report--

Long before the advent of the Internet of things, when computers were not personal enough and mobile phones, refrigerators and cars were not smart enough, do you remember when advertisements for a brand of oil filter were repeatedly shown on TV?

The price of a high-end oil filter is nearly double that of an ordinary brand, which is undoubtedly very expensive for users. As a result, this kind of high-end brand products encountered very serious resistance from users when they first entered the market. However, with its excellent technology, it gradually improves the performance of the car, while effectively reducing the probability of failure and avoiding expensive maintenance costs. So there is no doubt that people are more likely to change the oil filter than to change the car engine.

So the product gradually won the favor of users, and finally occupied the dominant position in the market.

The lessons of this old marketing story can be applied to some of today's misunderstandings and erroneous logic about data center infrastructure management (DCIM) solutions.

Myth 1: DCIMs are too expensive

Intel commissioned Redshift research to consult 200 data center managers from the United States and the United Kingdom. The biggest dilemma, according to the survey, is that 43 per cent of data center managers are still relying on manual methods to replace DCIM tools for capacity planning and forecasting.

As we enter the digital age, Microsoft's Excel spreadsheet software and even the traditional measuring tape still occupy a very important position in the toolkits of many infrastructure managers, which seems counterintuitive. But it is true that only 55% of data center managers now use the DCIM platform. The biggest objection to deploying DCIM is people's perceived cost, with 46 per cent of respondents saying they think DCIM implementation is too expensive.

At first glance, data center managers' concerns about costs seem reasonable. However, considering that DCIM tools can provide managers with effective information to identify operational problems and help determine the real operating costs, hidden costs, and the real causes of system outages, any cost-based objection to DCIM is misleading. You know, the 118 data centers surveyed can clearly quantify the average cost of each operation outage as high as $28900!

Don't get confused. Whether due to hardware failure, power supply, or cooling problems, large data centers inevitably face operational disruptions and downtime failures at some point in their lifecycle. Only those data center managers who have deployed DCIM analysis tools for capacity planning and cooling performance monitoring can clearly calculate the cost of business loss caused by operational disruptions, while only 14% of data center managers who do not deploy DCIM can do so.

In addition, since time is money, according to statistics, the average recovery time required for disruptions in data center operations is close to 8 hours. It is worth noting that 21% of the data centers with DCIM deployed can effectively control the recovery time to less than 2 hours. Only 11% of data centers that do not use DCIM tools can do this. The difference in proportion is nearly 2 times!

Misunderstanding 2: the implementation of DCIM needs more time and resources

Thirty-five percent of data center managers said that the manual approach to data center capacity planning and forecasting was due to concerns about the lack of the necessary time and resources to deploy more automated solutions. On the face of it, this reason seems very persuasive. But in fact, 56% of data center managers who use manual methods spend more than 40% of their time each month on capacity planning and forecasting.

Paradoxical situations such as data center managers reflect a vicious circle. Manual data center managers say they lack the time and resources to deploy DCIM tools for capacity planning and forecasting, precisely because they waste most of their time on tasks that can be done automatically by DCIM.

Misunderstanding 3: cabinet thermal sensors and spreadsheets are enough to easily complete the heat dissipation of the computer room.

The power consumption of the data center is very huge. According to the estimate of the National Resource Defense Council, the power consumption of the data center will reach an average of about 140 billion kilowatt-hours a year by 2020, which is equivalent to the annual generating capacity of 50 power plants, and the annual electricity bill alone will cost $13 billion. In addition to the power supply cost of server operation, heat dissipation also needs to consume a large part of electric energy, so effectively improving the efficiency of heat dissipation has become a necessary means to save the total cost of operation.

Fifty-seven percent of the data centers surveyed claimed that they had experienced heat-related challenges over the past year, which seriously affected operational efficiency. 63% of data centers are currently using DCIM analysis tools to help optimize cooling efficiency, while up to 20% of data centers currently rely solely on cabinet thermal sensors and spreadsheets for cooling. Data centers that do not use DCIM analysis tools are unlikely to conduct hot spot audits, let alone complete computational fluid dynamics (CFD) simulations.

The CFD function is the core function of the DCIM solution, which can continuously improve and adjust the cooling strategy and air conditioning selection through real-time monitoring information. DCIM systems with CFD capabilities will have a direct and positive impact on the operation of the data center.

Conclusion

Data center managers need to collect accurate data information, including energy consumption, heat, airflow, and utilization, in order to make correct operational decisions. With the popularity of DCIM tools and the increasing level of automation and control, data center managers gradually improve their operational efficiency and respond to common operational challenges in a timely manner, such as capacity planning and allocation, and cooling efficiency. Any cost-benefit analysis of DCIM investment should take into account resource savings, reduced downtime, and improved cooling efficiency, which will eventually have a positive impact on the return on investment.

In other words, the experience learned from the old-era oil filter mentioned at the beginning of this article is still applicable in today's digital age.

For more information about Intel data Center Management products, please visit:

Http://www.intel.cn/content/www/cn/zh/software/data-center-overview.html

Or contact us by email to dcmsales@intel.com.

The author introduces:

Jeff Klaus is the general manager of data Center Manager (DCM) solutions at Intel Corporation in the United States. At Intel, Jeff Klaus has more than 13 years of team management experience. His team is developing energy and heat management middleware products. Mr. Jeff Klaus holds an MBA degree in Business Administration from Boston University.

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.