📜 ⬆️ ⬇️

How four-hour support turns into weekly, and some features of HP, IBM, Dell paid service


We understand the intricacies of paid support for server hardware and correlate the cost of “sheepskin” to its dressing. It's about service support from IBM (Lenovo) , HP and Dell - companies are very actively promoting their own services in this area.


Paid support for the Big Three servers is called differently:



For simplicity, we will call them “paid support” or “service contracts”.


Memo - basic information about service contracts, their nuances and features

Basic information about paid support for HP, IBM and Dell


Paid support in general is simply the service of restoring failed equipment.


Usually it includes:


  • Certified equipment installation (may be required);
  • Remote monitoring, firmware update (may be required);
  • Receiving applications for repairs and maintenance with a predetermined reaction time;
  • Work on troubleshooting with a pre-agreed period and a lot of nuances;
  • Departure service engineer in your office, sending parts by courier service.

Any support package contains a combination of the above services, as well as an agreement on reaction time and recovery time . In addition, additional restrictions may be imposed, such as the geographical distance of your office from vendor service centers.


In this article, we will focus more on service contracts for entry-level and mid-level servers (HP Proliant ML \ DL1xx \ 3xx, IBM System X 3xxx, Dell PowerEdge R \ T), as the most requested.


In the tables, we did not indicate the cost of support, since by itself it does not carry much benefit and depends on the cost of a particular iron. To compare the efficiency of services from different vendors, we introduced the ratio of support cost to equipment cost.


The following resources were used as sources of information:



Basic paid support


Basic support is most common due to the simplicity and acceptable speed of eliminating failures. It covers both diagnostic work and replacement of equipment components.


The start time of work depends on the distance between your office and the service center, although not all vendors clearly specify the mileage. For example, HP 6 hours CTR are relevant, if the nearest service center is not more than 80 km. Otherwise, the recovery rate may be delayed for several days and incur additional costs for the engineer’s journey.


PackageReactionBeginning of work
on recovery
Availability timeCost of equipment support
HP Foundation Carenext business daynext business day5 days a week for 9 hours0.044
HP Foundation Care24x7N \ A24/70.372
HP Foundation Care - Call to Repair (CTR)2 hours4 hours24/70.671
Dell ProSupportnext business daynext business day5 days a week for 10 hours0.047

The reaction time can also affect the reaction rate. For example, in the description of support for Dell NBD it is stipulated that if you call after 5:00 pm, an additional working day may be required to process the application.


Have you noticed that there is no IBM in this table? The fact is that the company provides only the proactive committed service.


Support with proactive service


Packages with proactive remote support include an additional set of services and are usually more expensive than the previous category 1.3 to 6 times. In the table below you can see the main differences of this category from the base one:


OptionRemote diagnostics supportBasic support
Dedicated managers+-
Firmware and software updates, audit and performance forecast (several times a year)+-
Remote technical support++
Fast response to critical failures++

Remote access support is usually slightly faster in terms of recovery rate due to remote monitoring and, as a result, less time for initial diagnostics. In addition, some failures can be prevented in advance, and the manufacturer’s monitoring system looks after the state of the server park.


IBM service packages are similar in HP Call to Repair and Dell Mission Critical capabilities, that is, the time from remote diagnosis of the problem before the start of the repair is specified.


PackageReactionBeginning of work
on recovery
Availability timeCost of equipment support
HP Proactive Care Next Business Daynext business daynext business day5 days a week for 9 hours0.287
HP Proactive Care 24x74 hoursN \ A24/70.616
HP Proactive Care - Call to Repair (CTR)2 hours4 hours24/70.915
Dell ProSupport Plus NBD0.5 - 2 hours (additional service)next business day5 days a week for 10 hours0.115
Dell ProSupport Plus Mission Critical 4 Hours0.5 - 2 hours (additional service)4 hours24x70.306
IBM ServicePac0 - 4 hoursnext business day5 days a week for 9 hours0.283
IBM ServicePac - NBD0 - 4 hours48 hours (next business day)5 days a week for 9 hours0.283
IBM ServicePac - 24x7 8 hours0 - 4 hours8 ocloc'k24x70.964
IBM ServicePac - 24x7 24 hours0 - 4 hours24 hours24x70.444

For Dell, a separate Support Service is responsible for answering, which is divided into 3 categories:


  1. Standard;
  2. 24x7;
  3. Premier.

All of them allow you to communicate with Dell by phone or e-mail, receive updates and read documentation (Knowledge Base and other online resources), and differ in response time to your incident. For example, with the Premier category, a critical failure will react in half an hour, against an hour in other categories.


As part of Dell's ProSupport, you are also assigned a manager on the side of a company familiar with your hardware, allocate a separate communication channel, monitor the status of the server fleet and warn of failures, give useful tips for improving overall stability.


Remember that a proactive support package is usually sold with a certified installation service ($ 100 - 300).


Reaction time! = Recovery rate



Studying the input plates, you have already noticed two different concepts - reaction time and recovery time . Contrary to simple logic, manufacturers of server systems tend to separate them and introduce some confusion in the descriptions of paid support.



In practice, the reaction time looks like this:


“Thank you for your message! We regret that errors have arisen with the equipment, but we will definitely fix everything. Wait for the response of the engineer »

In the basic assistance packages you can expect a day - another, during which you will be left to yourselves and to the disgruntled management. To get some guarantees, you can choose a more expensive package with a specified recovery time.


But here, not everything is so smooth, because the following factors come into effect:



“We ourselves regularly communicate with the support of vendors, so there is some practical experience in this regard. For example, HP and Dell work the fastest - parts were usually brought to us the next day. Among the shortcomings of HP, it is possible to point out only a rather confusing application registration system, with several stages of confirmation ”

In addition to all of the above, it is worth remembering about the additional requirements for advanced packages like HP Proactive Care or Dell ProSupport Plus . Support conditions for them include requirements for a certified installation ($ 100 - 300 per server) and remote access. Of course, this is not so scary, unless you have a state or security agency.


Partners provide service support, not manufacturer



Any service request that requires the personal presence of an engineer goes down to your nearest service center. Often this is the largest system integrator in the region from which you just bought the equipment. Specialists of the manufacturer visit only very large customers with rare systems like the HP Superdome .


“The first seditious thought: once a specialist comes from the same company where the server is purchased, is it worth ordering additional support from the manufacturer? Very interesting question, and we will answer it very soon. ”

In general, both internal and external repair and maintenance engineers have a similar knowledge base and the same set of certificates from the vendor. For example, for repairing IBM System X or HP ProLiant server systems, at least the following qualifications are needed:



See how many requirements you need to fulfill to get them.


Accordingly, any engineer with a similar set of knowledge can come on behalf of HP or IBM to assist. In this example, there is not much point in checking the competence of an engineer, because the vendor does not send service requests to anyone.


More interesting things with the service from the seller. For example, when you buy a restored server, you can be offered your own warranty with a visit to the site and remote assistance. How to make sure the quality of support, if the vendor does not manage it?


You can go two ways:



It turns out that in terms of technical knowledge, the specialists of the seller and vendor are likely to be on the same level. However, differences in the scope of services for service contracts will be significant.


It is impossible to say unequivocally “buy from N and use their guarantee” or “order servers with proprietary support”, because in each particular offer the balance of pros and cons will be different. To make your choice a little easier, we studied a lot of specifications of HP, IBM, Dell and prepared a pivot table for their packages (see the spoiler at the beginning of the article).


Distance matters



We live in a very large country, where you can go from one end to another for a week. Of course, this could not affect the timing of the provision of technical support, even for money. All manufacturers of the “big three” have a limit on the distance between your office and their service center, which directly depends on compliance with the agreed speed of recovery. But not everyone publishes such information explicitly.


For example, the HP 6 hours CTR (from a large table in the sidebar) are relevant only when the distance to the service partner is no more than 80 km. If you fall into the interval of 80-160 km, then the delay increases by 2 hours. With more distances, we will talk about days and need a separate discussion of this nuance when buying.


“With IBM's support, the situation is ambiguous. For example, we waited for about a month for one cable to connect the second basket to the server, but it was not the right one. The bureaucracy at all stages of the application process is also strained to problems with logistics. It was an unpleasant surprise for us, the inability to quickly reach the engineer. After the series “your call is very important” and “wait, now I will connect with the right specialist” you will be switched to a competent employee, but the problem will have to be repeated many times and for everyone ”

In addition, all the stated dates refer to the situation when the necessary spare parts are in regional warehouse. If this is not the case, then a legitimate increase in four-hour support to two days is possible, for example. And what will happen if the components for replacement are only in production in Europe, I don’t even want to think.


All these nuances with distance should be checked by the manager who sells your equipment with support. But the human factor can enter into the matter, and then you will find out about the impossibility of getting paid service only at the most inappropriate moment. In this regard, only customers of an individual configuration (from the factory) are more or less protected. Since the application is sent directly to the manufacturer, its specialists will certainly check not only the hardware compatibility, but also the possibility of providing the service.


Responsibility for service before the law is the seller



The last thing you want to think about when ordering a new equipment is the poor-quality performance of its warranty obligations by the manufacturer. If in terms of equipment we are somehow protected by the law and the established warranty process of the vendor, then the service is different.


The fact is that service contracts are sold as a commodity. And the seller is responsible for compliance with the expectations before the law. The manufacturer here plays the role of a certain “arbiter” who can somehow influence his partners. But all this is done solely at his discretion, and you can hardly put pressure on the manufacturer.


“But even if you don’t think about very extreme options like a lawsuit, then the question of the expediency of acquiring a service from the manufacturer involuntarily sneaks in, if the work on it will still be done by a partner.
Do you remember that the service partner and the seller are usually the same company? ”

Before answering, let's imagine the situation that you ordered a three-year service with a response time of four hours. That is, they gave a lot of money and expect help in a critical situation. And then the company that sold you the equipment goes bankrupt.


There are 2 options:



Thus, support from the manufacturer can be viewed as “service virtualization” without being tied to a specific artist. At the same time, the seller’s service can only be sufficient if we are talking about a stable company with an extensive network of representative offices in Russia.


Extended support is also available for refurbished equipment.



Due to the specifics of the business, we can’t get past the issue of support for remanufactured equipment (not to be confused with used ones). The fact is that in matters of support, the vendor considers officially restored equipment as “practically new” and allows you to purchase an official service contract in addition to the standard year of warranty.


However, the purchase of refurbished equipment usually means a desire to save, including on support. So why the seller does not offer their own warranty? At the same time, you can make the recovery more operational and add a couple of “trumps” that are not in the expensive support from the manufacturer.


For example, we managed to achieve the following conditions of a free warranty for equipment sold by the Network Administrator :


Server typeReactionBeginning of work
on recovery
Availability timeRemote diagnostics and repairGuarantee period
Newup to 1 hour
(usually 5-10min)
24 hours with spare parts, 48 ​​- with their absence5 days+5 years (3 years from the manufacturer + 2 years own)
Recycled (refurbished)up to 1 hour (usually 5-10min)24 hours with spare parts, 48 ​​- with their absence5 days+3 years own

As you can see, the reaction rate is even more interesting than many paid options from famous brands. For example, in 2 days you are guaranteed to receive a replacement component, even if we have to personally go to the supplier in the neighboring town, buy it twice as expensive and send it by emergency delivery.


“Do we often change servers in the collection? Only once was this, one of the customers in Izhevsk could not connect to the local console of the server. Moreover, there it was somehow strange - our server found remotely our monitoring and did not detect any problems, even the iLO console worked. But the problem needs to be solved somehow, and we just sent another server with a courier.
And so, each piece of hardware is checked very carefully, to avoid "

All this is possible thanks to :


  1. A separate support unit that is not distracted by sales and design work;
  2. Work only with the largest courier services that can provide delivery of goods to any part of the Russian Federation in 1-3 days;
  3. Special conditions for ordering parts. Thanks to partnering with HP, IBM and Dell regional distributors;
  4. Emphasis on the quality of the supplied equipment and its service support.

Let's sum up


The basic set of service contracts for IBM, HP and Dell is generally the same and may include additional services, such as remote proactive monitoring.


Is it a rather expensive service with a new server? It is important to assess the criticality and allowable downtime for applications that will run on it. If you once compiled a table of correspondence of business services to their influence on the company's activities - it's time to remember about this work.


We will issue our recommendations and thoughts in the form of a small summary:



')

Source: https://habr.com/ru/post/308830/


All Articles