Listen to the full podcast from Thriving at the Crossroads podcast with
Welcome to the Thriving at the Crossroads- today, I’m pleased to introduce Tyler Constable from syslink, a Swiss start-up. Tyler is the Director of Sales Engineering. I realize you are a techie today, so we are going to have some fun today! I don’t get hardcore techies that often. My geek socks will be going up and down the whole interview. Let’s get started- why don’t you describe the problem you solve?
First I’d like to take a step back from when you think of SAP. When you think of SAP in its simplest fashion for what it’s doing, it’s helping to run Sales, Logistics, Production Planning, Financial Operations, any aspect of a business. If SAP goes down, Sales are not occurring, production is not being produced, product is not leaving the door, pretty much the business can be at a standstill. This puts immense pressure on the technical teams supporting the system.
Making it worse, SAP is not your standard application where the health of the system is regulated by a simple up/down. There’s a long list of checks that need to be manually combed through every day. And that’s where it gets really techie and geeky looking through those individual checks, just making sure the system is running optimally and that it is secure. This is extremely time consuming. I actually came from an organization where we did management of hundreds, actually thousands of systems and it took us over 100 man hours a day to comb through the systems to make sure things were running securely and optimally. But if it’s not done, it leaves the teams in a reactive mode and troubleshooting becomes a total pain in the rear!
I think that’s a key that as you highlight and talk about system functionality- keeping uptime is crucial. If sales order are not going out the door, if we aren’t recording inventory movements, etc. The lifeblood of your company is bringing cash in the door. You end up in a world of hurt really fast in terms of what can happen. How big is this problem- any horror stories you can share? Are you seeing a lot of companies having problems with downtime? Let’s talk about the product and why you exist.
Let’s talk downtime first. We’ve spoken with global companies with many modules- that state if the system is down it can cost them 10’s of thousands of dollars a minute. Not just by the outage, but by the minute! We have one customer from a prior life where a production instance on a landscape went down and it lasted for 3 hours and 20 minutes and cost the business millions of dollars, and some people their jobs. That’s real serious- this is obviously an extreme measure and was a multinational conglomerate, but the financial impact scales with the size of the company. That’s just the example of a production down. We also have to look at the performance of a system. If end users are having a half second delay just running through the system, then you compound that by thousands of users, that’s a lot of time wasted throughout the day.
You talk about performance of the system. We’ve all been there when it’s “hit the execute button on the transaction and go get a cup of coffee” because it runs so long. Performance is a huge deal, and it can really add up. Talk a bit more about how you guys help a company in those situations and the health and system checks?
There’s a lot of things that we do. We’ve been talking about performance. Our product Xandria reaches out into other areas as well. If we think about those manual checks that these individuals have to go through and the technical teams have to work on. We take all that off their plate and let them focus on more productive tasks such as SAP projects.
We know SAP projects are not just click “Next” five times right? These are large scale projects. We don’t want to have people reviewing to make sure the system is secure and running optimally. There’s automated solutions for that. By providing them with real time health data and analytics, Xandria gives the support teams a single pane of glass for all their system where any soon to be issues can be identified and resolved before they become business impacting.
This strikes me as reminiscent of another interview I had done a few weeks ago that was, a little more under the covers and not directly exposed to the business, but is so critical to the health of it. It gets IT out of the business of putting duct tape on things and putting the basic components in place. They get out of being the mechanic, and helps them be the strategist to actually look forward on those projects. It sounds like this product is something similar, would you agree?
Absolutely, the whole idea is that your administrators- we want them to be thought leaders not doers. We want Xandria to be the doers and for them to absorb that information to make sure that issues don’t happen in the future or to go through projects.
Okay, so for those that are a little techie, let’s talk about the product itself and what it does and how does it work? Let’s go into some more details there.
Sure- usually SAP has its own monitoring system called CCMS. We don’t want to touch CCMS- so we embed ourselves directly into the product. So- directly into SAP, directly into the database. And we are pulling information in a way that we see best fits. The product was designed by basis consultants for basis administrators. It’s not just some monitoring company that says “Well we’re going to get in the SAP world now”. This is specifically designed for SAP. It’s not just the standard monitoring tool. When you think of monitoring tools, I’m watching the network speed, making sure that my hard drives don’t crash or anything like that. Xandria’s more than that. We actually have some components in there than can do predictive analysis. That’s all done automatically by the system. The system automatically is looking at trends of maybe its CPU or file system space or tablespace if it’s Oracle and its estimating when system capacity may become an issue if the trends continue at the same rate they are growing right now.
So you’re building in a little predictive analytics as part of it to really help people be proactive in their planning as well. At the current growth rate of my database I am going to run out of space in XYZ days, months, etc. Is that the general idea?
Absolutely. And we actually give visual graphs so you can actually see it go off into the future. You can identify if I have a week to fix this or 3 minutes! It gives the support staff time to allocate when they need to fix what issue.
It tells them if their hair is smoking or just on fire? Sounds like it could be pretty helpful. Any time you have a visual display- that’s often helpful. It’s the old adage- a picture is worth 1000 words. Tell us some success stories of how it has helped some of your customers.
If we look at some of our European based customers we’ve got some of the largest retailers out there in Europe that are running our product. They’ve come in with large scale issues where production down has cost them hours and hours of lost data. There’s credit card transactions that are not being pulled- I can’t remember how many per second, but if there system goes down, that’s all data that’s being lost. All future BI analytics around that person that bought that Chapstick and what they might want next time. All that data is very valuable to them. If it’s gone or lost or even if they can’t pull the data that fast. If the speeds are not correct- all that gets lost.
There’s definitely a lot of advantages to not only the system monitoring and predictive analytics but also in notifications. How do we know about that case where SAP is running slow for that customer and they are not able to pull every time someone runs a credit card. Who’s being alerted of that? We actually have smart notifications. We’re all human and we all have lives. I’ve always thought it was funny- we can set up text messages, but what do I do if I am at my kid’s soccer game, or maybe I’m at a wedding? Our notifications are set up to notify the first person and if they don’t respond within a period of time, its going to notify a second person or group to say, “I’ve sent this notification out and no one has responded, please help this is going to be a problem!”
So it sounds like it has some built in escalation procedures. We’ve been in IT forever, right? I’ve been in IT pretty much my whole career. Who hasn’t had the phone call or had to track someone down on a weekend. At the kids’ soccer game, you’re at a movie, you’re wherever. How many levels of escalation can someone have? How many levels do companies normally have?
What we usually see is an individual on call, and that’s the one person, right? It goes off to that one individual. But should that individual not get back, it then blasts out to a group of individuals. So we usually see for the most part just two escalation levels- it goes out to that one person, and then if they can’t respond it’s all hands on deck and then someone will get on it.
So two is usually the number of levels. First the individual and then the peer pressure kicks in- “Someone’s got to answer this, right?”
Absolutely. If you think about, notifications are usually going out when something is an emergency, right? So if you’ve got a plant set up where it’s escalation point one goes off, and then we wait 5 minutes. Then escalation point two goes off and we wait. The next thing you know we are 20 minutes into the issue and no one has responded yet. We want that second notification to go out to everyone so they can get back and respond within a five minute period of time.
Taking it back to your retailer example, it can thousands of transactions before you blink, especially someone that does some more volume, like a Target or a Wal-mart, something goes wrong in their system, you have to respond really really quickly because of the escalation of volume of transactions. Any stories you can share about averted disasters or unusual problems- I like the war stories, because it really demonstrates how products can make a difference in the average human’s life and the company’s life.
Averted disasters is the name of our game, so…… I’m trying to think of a good example. To be honest, I can’t think of one off the top of my head.
You’ve avoided too many disasters. What, too busy wearing your superhero cape?
Yeah, you think of- if there is a problem with SAP, the majority of the business may come to a crashing standstill. That’s the superhero story every single time.
Are there particular sizes of companies that tend to look at the services you have? Do you run the gamut of smaller companies? Sometimes you see companies of $100 million or less on SAP all the way up to $100 billion. Is there a sweet spot for you guys? Are there particular industries that are more appealing or tend to look at these services more than others or is this something everyone is getting in on? Tell me a little more about that.
The product is actually built out where it can handle one system if we wanted it to. But the system was actually built for a managed service provider which is going to be handling hundreds if not thousands of systems. If you want to get the true value of the system, it the companies that are…. Say the large auto manufacturers.It’s the huge SAP landscapes that Xandria’s built for. It can do the small ones, we don’t disregard those by any means, any SAP systems are important. But it can handle the Goliath’s, the hundreds of systems- that is what it is built for.
So you are built for the companies that may have one global instance, to the companies that are regularly buying and selling other companies and have 50 instances, because we run into acquisitive companies. So it sound like you are for anything soups to nuts in terms of what their SAP infrastructure looks like. One system, 50 systems, doesn’t matter. Is that a fair statement?
Absolutely- you brought up acquiring and divesting too. There’s companies that are always buying and selling each other, right? Parent companies with holdings, and so on and so forth. I’ll get a little more technical here. When we are pulling in different systems, we have to put in agents. The product is actually smart enough to know when you bring in a new system. We tie the system to Xandria. Xandria is smart enough to then jump in and recognize the type of system it is. It can say “Hey I recognize this is a BI system or a BW system therefore I am going to assign these checks automatically”. All that manual interaction is done. So if you’ve got a landscape where systems are coming and going, Xandria’s great. You just tie it in once and it automatically goes “Ah, this is a PI system. It needs these checks.” The thresholds are automatically set and its really just the administrator going in and fine tuning the thresholds. Everything else is automatically all brought in. And on top of it, it is all searchable. You could say, there’s a security release for any systems that is 730 or higher. You can just look, see the systems that apply, and you can see where you need to hit your security notes. So it’s built for companies that have systems coming and going.
I hadn’t really thought about your peripheral, your PI’s, your BI’s- that’s a really great point. It’s not uncommon to stand up different systems. We need a prototype system for that, whatever system for that, we changes paths as we do patching, all these things we have to do from a technology perspective to keep the business moving forward and execute the project. It sounds like you are the easy button. When you stand up these new instances, systems, boxes, sandboxes, etc. It recognizes it and does those proactive checks for you. That way you get away from that ugly little, “Gee, is this a productive system wasn’t set up the same way as others”. Or you find checks that are turned on in one system but not in others. It’s the inconsistencies, we never run into those, right? And we find things work in one system, but then not in another because they aren’t set up the same and say AAAACK! I just need something I can replicate!
It’s that human nature getting in the way again- again, it foils us again!
Yes, curses for our own natures! That is our very humanness right? The ability that we miss things as we set things up- these can actually lead to serious problems in terms of how systems are set up or maintained. Now, you brought up that security aspect, that big old ugly elephant in the room. For more and more companies it is becoming a challenge especially with some of the breaches we have seen, various governments and hacking making the news. We know companies have to protect their Intellectual Property. Talking about security and patching- what’s in the tool around those types of activities?
There’s a whole world of discussion around this. Security is a hot topic right now. When you look at things from a very high level there’s network intrusion, we’re not getting involved in that. Then there’s SAP security, who has the right roles and profiles, and we’re not talking that either. But we’re talking about change control settings in SAP. In a normal SAP landscape, you want to all your development in Development, test it out in QA, and push it into production. Therefore your production system should never have any open change control.
Open production client- no! We don’t want this.
Absolutely. It’s easy intrusion. And it sounds so comical, right? But you see it all the time. Someone for example will flip on a switch to make a quick change in production whether it is allowed or not allowed. They pop the client open, make the change, and go get a quick cup of coffee. Six months later, they forgot that they opened up the client. Then something terrible happens, or someone just stumbles across it. Or maybe nothing bad happens, but an auditor comes across it. And then you get into the whole game of everyone’s pointing fingers because no one can remember who did what. With Xandria, we can watch all that. We can notify when, uh oh, someone sets a client to automatic changes without recording. Then all of sudden, that gets notified and maybe goes out via text messages to managers, emailed to the internal auditing team. Or maybe there was a legitimate reason for it. We can actually put a change control document into that incident or case for tracking. Therefore, when the auditor comes in at that horrible time of the year and look at everything, they can see it. You can sit them down in front of Xandria and say this is all the times our system has or has not been opened. And it will also show the reasons why and who it was. So if Bob’s over there in the corner and he’s trying to do something malicious, hopefully not, but it’s actually going to be tracked that Bob is the one that locked this up or unlocked it and he did or did not lock it back up again. That’s one part that is huge- just the notification component, but also the auditing. For the auditors, it is “here’s the system, talk to you later”.
Let’s be honest- we do still have to on occasion open the client. Things happen for different reasons, but I still run into this. The only way we can make certain types of changes is to open the client. And you’re like, “Uh, okay”. But then you are right- what controls and what approvals. The ability to track that, so that the auditor can see what the change was and why- and it’s all documented. Next! Moving right along…….. in terms of the audit. Otherwise, you are digging through emails, change control documents, or at worst, if it wasn’t well documented, someone is scratching their head trying to remember why. So it sounds like an auditor’s dream in terms of the systems and what is set up?
Yes, definitely, and that’s where we hear from our customers. I think the notifications are cool, but we actually hear from our customers about the audit functions and hear positive feedback. One of the other aspects that is kinda security but kinda not is looking at the different system components. Again, there is always the case where you are working in development and maybe you put in some support packs and then you go into quality and you put in some support packs and something happens and you get pulled away. Now you start working on a different landscape. Then you go back and are moving transports. They are fine in development and in QA, and then you get to production and something happens. Now boom- everyone is scrambling trying to figure out what is going on. Is anyone looking at the different versions and are they all aligned. The system is actually pulling that at all times. And it notifies when there was a change. Not that someone is going to do surprise support packs, I don’t know who would do that. But when there are changes to levels, that is all being tracked and watched.
Sometimes you get surprised at how things can end up out of sync. It happens, right? It’s interesting that you can record those differences. So it sound like not only is it for the basis team, but also for the analysts and others that may find inconsistencies trying to figure out if something is different between my systems. So it’s not just the Basis team geeking out and hitting the easy button.
One other cool thing, sorry, I get excited about security. The way SAP works for the non-super technical geeky audience- when you go in, there are different security changes. Tracking in Xandria that allows us to monitor specific changes that can be implemented in SAP that require a system bounce. We are able to monitor those changes before a system bounce and provide notifications. Catching a security breach after it happens is wonderful, but we want to catch it before it happens. We want to catch it before the change is even in place, that’s our goal.
You want to stop it before it happens, if at all possible and know something is going on with as much notice as possible. Because catching it after it happens is like the proverbial closing of the barn door after the horse has bolted. Great, the door is closed- but you want to stop that in its tracks. Let’s talk about your customer base. You’re more established in Europe. I like to ask my guests about their customers, since I interview companies that are just acquiring their first customers (Alpha, Beta, Customer) to those that are already established with more than 5 customers, category E for existing customer base. Where do you fit?
I’ll answer it kind of comically. We’re calling ourselves the 20 year start-up. Our group in Europe, they have over the course of the last 20 years acquired close to 250 customers. We’re just now trying to bring this into the US. If you count just the US part, we’re definitely in category A. If you consider the whole company, we are definitely E. So to help answer the question, we are a twenty year old start-up.
I think that’s category E given the sheer base you have in Europe! You’re just expanding into the US. The product is no different for US customers, is it?
It’s the same product, just bringing it over to the states.
All right. At this point I want to shift gears a little bit and ask you a fun question. I love to travel, as do most of my podcast guests. Tell me your favorite travel destination you have ever been to and why it was your favorite?
I’m a fan of Keystone in Colorado for snowboarding. I’m a huge fan of Keystone- I try to get out there at least once a year. Hopefully get out there this year and get a little snowboarding in.
Fantastic- thank you so much for joining us today. I appreciate your time and sharing a little more technical topic today and how we can help the Basis team’s hit the easy button for their infrastructures. Thank you so much Tyler!
A Large Retailer Increase System Efficiency with Over 120 SAP HANA Monitoring
It's REALLY hard to get a clear visibility of a complex SAP system performance.
Or is it?
4 SAP Monitoring Challenges You Must Solve
SAP software runs every aspect of many enterprises, from mission-critical line-of-business systems...
5 Agent-Based vs. Agentless SAP Monitoring Myths Debunked
“Our IT uses agentless monitoring solution, why can’t we use it for SAP?” Do you hear that often?...