Automic Workload Automation Review

Handling SAP processes is very easy, while distributed architecture keeps jobs running


What is our primary use case?

We use it for assisted process change, and we are using it for basic operating system-level UPROCs or jobs, and there are certain jobs that it runs for the net backup.

How has it helped my organization?

The solution has been here for the past ten years, there is a definite business value-add; the batch/shell scripts running in the environment can be controlled centrally, SAP Processing; Backup Jobs and many more with no or minimal interventions.

In general, in any environment where there are more than 500 or 600 servers, each server will have settings and scripts doing their jobs, moving files, etc. There may be a bunch of scripts that run in a workflow. If you don't have a centralized tool for workload automation, it becomes problematic down the line because, as the environment grows, as IT grows in any organization, the number of scripts grows accordingly. If you have a centralized workload automation tool, you can completely control such jobs, or file transfers, or any job that is critical to a specific application/server. So it provides ease in handling scripts.

Also, it helps with manpower. If you have server admins to take care of those scripts, you need more admins, of course. But if you have one such workload automation tool, a single person can control, monitor, and see the behavior of the scripts in the environment: How well they are running and, if they are failing, which scripts are failing. That's the business value-add that I see in having any workload automation tool, like CA Dollar Universe, which is the one we have here.

What is most valuable?

The most valuable feature is the one for SAP batch processing. It's not just ordinary job processing. There are certain other mid-level workload automation tools which can handle the OS level, but SAP is something which is really very critical. Automic stands out from the ordinary tools because handling SAP processes is absolutely easy with it. Integrating SAP applications with Dollar U is very easy. It's just a few considerations and there you go. You can initiate your processing.

What needs improvement?

We are currently at version 6.7.41. One improvement area that I can see would be a centralized licensing part. I've heard that has been already taken care of in the latest version. I'm not sure how true that is, but that's one thing that should be there: centralized licensing.

Another issue is that at times there are certain jobs that are triggered one after another. It would be helpful to have a more user-friendly way of seeing how those jobs are connecting from one server to another. Suppose there is a workflow that is running between ten and 15 servers. It's always challenging to figure out which job is connected to which job on which server, for a newbie, if you haven't designed it. That has to be more user-friendly where you can see the complete workflow of a process or a job.

For how long have I used the solution?

One to three years.

What do I think about the stability of the solution?

I haven't seen any issues regarding stability.

What do I think about the scalability of the solution?

It's very scalable. It's easy. You just add on the resources and you are there. The management server doesn't get loaded up, it doesn't have to process anything. That makes it cheaper as well. Scalability is pretty easy.

How is customer service and technical support?

On a scale of one to 10, with 10 being highest, I would rate technical support a nine. So far, I haven't gone unanswered for any of the queries, except one. Their response time is pretty fast. It depends on how severe the case is. If it's just a general query, they respond within a day. If it's really critical, where your business is impacted, they respond within half-an-hour or an hour. I have had a really good experience with the tech support.

Which solutions did we use previously?

I have been an admin on other tools as well. I was a contributor to and implemented BMC Control-M.

How was the initial setup?

I haven't done the initial setup, but I think it's pretty straightforward from what I have seen in the documents. My feeling is it should not take more than an hour or two to get it up and running. If everything is ready, your database is there, and you have the right amount of resources on the server, it shouldn't take more than an hour or two hours.

In terms of an implementation strategy: 

  1. You should have a database. 
  2. You need to figure out what components you're going to go for.
  3. You need an estimation of the number of jobs you are looking for to plan out the resources on the server. 
  4. Finally, you need to think about how you will roll out access to the users: a thick client or a web console. 

Those are the things that need to be factored in before beginning the installation. The accessibility part can be dealt with later, but the resourcing of the database on the server and the management server have to be spec'd out before.

What's my experience with pricing, setup cost, and licensing?

I understand that AWA is cheaper than Control-M, but I'm not certain about the numbers.

Which other solutions did I evaluate?

What makes it stand out from the competitors I have seen is it has distributed architecture. If you look at BMC Control-M vs Automic Workload Automation, the brain is the central, enterprise management server. That's where all the jobs reside. Every day, a new set of jobs loads onto the agent, and then the agent executes. If the central server is down, there will be no jobs executing across the environment.

However, when it comes to CA Automic, it has a distributed architecture which means all the logic, all the jobs, reside on the agent itself. Irrespective of whether the management server is running or not, your jobs will execute in a timely way. The only challenge could be that you will not be able to see their outcome. You will not be able to monitor them. That could be a challenge. But again, at least the jobs are executing in a timely fashion, as they're supposed to, in your environment.

What other advice do I have?

It's the same for any end tool we implement: Be clear with the requirements. Apart from that, everything is pretty smooth and straightforward. You can look at the tool and understand where things are going. There is no rocket science that you need to be worried about. But you do need to be aware of what you're doing.

Regarding the number of staff for maintenance, it depends on how exactly you want to maintain it. We always keep all the UPROCS, all the jobs that we have in an environment, on a centralised server as a backup. The maintenance is up to the individual organization, how robust or how limited they want it to be on the day of a crisis.

In our organization, we have a team of nine people handling the tool. We have more than 12,000 tasks that are scheduled to run each day, and more than 100,000 job iterations happen every day. It's actually a really big environment. We have more than 1,400 nodes connected to it, and we are bringing in 300 more. At each of those additional nodes we are expecting four to five jobs. So that will add about 1,500 tasks. The number of iterations expected is still unknown.

Right now, we execute jobs in three regions: Europe, Asia-Pacific, and America. We are only using AWA in the European region. We are taking it into Asia as well. That's the next expansion of the tool.

The admin roles include handling new requests for creation of the tasks and sessions, as well as the changes to existing jobs, including notification, and daily scheduling. In addition, there is the daily maintenance part. We check for jobs that are failing every day, why they are failing, and we will try to mitigate the problem. It could be the agent needs to be purged, or the agent is not running, or the credentials that were given for a specific job are not there anymore. Those are the sorts of checks we do on a daily basis to keep it healthy.

I rate Automic Workload Automation at eight out of ten. What comes to mind when I consider that rating is the distributed licensing, that every server has to be licensed individually. The second is the workflow of jobs connected on multiple servers.

Disclosure: I am a real user, and this review is based on my own experience and opinions.
Add a Comment
Guest
Sign Up with Email