DCRUM - Application insight into the darkest corners of your DC.
Updated April 27, 2017

DCRUM - Application insight into the darkest corners of your DC.

Ulf Thornander | TrustRadius Reviewer
Score 10 out of 10
Vetted Review

Overall Satisfaction with Dynatrace Data Center Real-User Monitoring

Dynatrace Data Center Real-User Monitoring (DCRUM) enables quick insight into the status and performance of all transaction based applications and some others too, like Citrix and VoIP.

When things go wrong, a quick glance reveals what part of the application delivery chain isn't performing as it should. Automatic baselining of all traffic also helps separate out deviations from the norm, such as new releases of code going wrong or design flaws in the data flow. Many solutions are covered out of the box and the ability to support inhouse protocols for legacy applications is important.

  • The ability to correlate Citrix users and their performance to an actual backend application.
  • Full insight into SAP over all the protocols as well as the customized code makes DCRUM a nice augmention to Solution Manager.
  • The high capacity to decode 20 Gbps traffic on a single box/probe makes it easy to slot in even in high density populated DCs.
  • In an effort to help understand and decode unknown traffic the configuration wizard helps and suggests what things to use in the datastream, since hardly any customers have documentation that covers the network traffic/protocol.
  • It's also very handy to be able to define specific traffic to/from a node as Software Services when they are a shared resource, such as a database, web farm or a DNS. That enables you to cherry pick from as many software services as you want when building the application delivery chain which makes up what a user perceives as "an application".
  • High capacity netflow processing helps you to get insight into devices that might be out of reach for various reasons.
  • One very convenient thing is the self-monitoring/maintenance of the system. Often the installations are left running for a very long time and there is comfort in knowing that they will maintain themselves with data rollup and cleaning tasks so you won't be met by a blank screen due to a full HD or the inability to restart without user intervention.
  • Major upgrades process is sometimes unpredictable.
  • The use of SQL Server should be evaluated for something else.
  • Easier SSL key handling.
  • It has reduced the number of war rooms as well as the number of people involved to address issues.
  • It helps in utilization trending for network capacity.
  • It has prevented poor solutions from hitting production.
  • When the various business units launch own initiatives such as third party tools or new platforms, it's become extremly easy to detect.
  • The reports help in the field of continuous improvement as any changes are immediately discovered and can be compared to history. Deviations from what you have decided as a tolerance corridor can be used to trigger alarms, both positive and negative.
There are a number of similar products in the field. Their challenge is that they are usually very hardware demanding or very narrow in terms of what questions they address as well as to what solution/protocol they can work with. Most of them are not so flexible as DCRUM and their support is traditional whilst the DevOps lab at Dynatrace is very agile and quick to respond.

It works really well on all things based on HTTP and the specific decodes it supports. There are guides for configuring but they are not "hard" as you can tweak and twist the decode with RegExs .

For the other decodes included, the same applies. You also have full freedom to draw reports backwards and forwards as all data is exposed in the DMI (Data Mining Interface).

It also sports a nifty Smartphone interface where you can control what is being published in a role-based setup, making sure your managers only get what they ask for and are spared the techie details.

Dynatrace NAM (formerly DCRUM) Feature Ratings

Application monitoring
10
Database monitoring
8
Threshold alerts
10
Application performance management console
10
Collaboration tools
4
Out-of-the box templates to monitor applications
10
Application dependency mapping and thresholding
9
Virtualization monitoring
9
Server availability and performance monitoring
10
IT Asset Discovery
3

Dynatrace Data Center RUM Support

The only thing missing for a 10, would be if they texted or called you back.
ProsCons
Quick Resolution
Good followup
Knowledgeable team
Problems get solved
Kept well informed
Immediate help available
Support understands my problem
Support cares about my success
Quick Initial Response
None
Yes - There is a premium support available that gives you a dedicated phone number and a named handler. This is probably an option that appeals most to customers that have global apps that need follow-the-sun support.
Yes - 

Yes and no. Serious bugs can be solved in a few hours with a personal patch while "grey zone" things or things of less impact might take more time.

These things can be how a decode interprets certain content. This has improved in the later releases though with the LUA based decode that are far easier to tweak than the older decodes.

Several times the Product Mangers has offered and taken over the dialogue directly with the stake holders to precisely understand why they request a particular function on report to cover a specific use case or business need.

Half of them time this leads to something improving in the product and the other half leads to a more educated stake holder :-)

Using Dynatrace Data Center RUM

Best of breed.
ProsCons
Like to use
Relatively simple
Easy to use
Technical support not required
Well integrated
Consistent
Quick to learn
Convenient
Feel confident using
Familiar
None
  • Any kind of network based statistics.
  • Any kind of network based discovery.
  • Application response time versus network response time.
  • Device discovery and baselining.
  • Customized reporting.
  • The dependeny on a SQL database forces you to make decisions on what data to keep for a particular period to prevent DB growth in a bigger network.
  • If there is no DAN (Data Acquisition Network) or previous experience of using TAP's or other means to get to the packets, the initial setup can be timeconsuming, but this goes for all similar products relying on a copy of the network traffic.
Yes - Slick - it gets the message across of the current status for applications and network. No access to minute details but that's not expected on a small screen.