The case for blade centers

When it comes to blades you have to realize of couple of baselines. The first is that you must understand the value of virtualization. The second is that you have a long term view of your world.

So now that we have established the value of the baseline, you need to understand the two main types of blade setups. The first is many applications on one blade center and the other is many blades servicing one application. The latter is way kewl and is what we use to call “grid computing”. That buzzword has fallen by way of the “cloud” and other similar useless marketing technical terms. I will focus this post on the many applications on one blade center. Please read my post on consolidation. I am even using the same graphic for it is that similar. I am also lazy and graphically challenged.

In the past when you had to replace a server or had to buy a new server for new applications you were forced to use a crystal ball. Yes there were many tools to aid you but in effect you had to estimate what you were going to need for several years. In the first years the computer is oversized and underutilized over time the server starts to get utilized as per the requirement and without due diligence the server will become overworked and over utilized. The roller coaster effect takes over. This is the reality faced by most organizations.

The advent of blades allows you buy as required and expand as needed. The initial costs of the infrastructure are high but you can add server instances in places where it would have been tough to get funding. Overtime the cost per server is dramatically lower. Do not underestimate the value of this. It is also a caution to prevent what is termed “server crawl”. The proper implementation of blade infrastructure will reduce more costs than just the hydro and air conditioning that everyone is talking about. It gives management and extreme flexibility and a pay as you play approach.

By using a hypervisor you are able to adjust the supply to the demand without the long processes of procurement and implementation. Benefits of backup and system management now become more manageable as well. The big win, regardless of the marketing hype is that you can actually think big, start small and scale fast.

On a vendor note, be careful of the hype. Deal with vendors that know data centers and that have blade technology for your environment. Some manufacturers are for the many blades to one application model and others are for smaller shops that will not expand beyond the one blade center. Know your business problem before you buy and commit.

Onboard computing realities

Onboard Computing

I was recently asked by an onboard computing company to be on a customer panel to answer questions.  They provided all of us some questions to answer and this post will share with you my answers.

Before the questions can be answered several things the reader of this post needs to be aware of.  First what is onboard computing?  This is actually one of those simple questions to answer.  It is a device attached to the engine system bus of vehicles.  This device collects engine information in real time and stores it for future retrieval.  This retrieval can be immediate or a store and forward approach.  These vehicles in my case are large tractor trailers and our straight truck fleet.  We currently have over 800 of these units rolling around North America.  In fact I have more rolling infrastructure then static desktop computing.  Also, the computing power and complexity is higher on my rolling fleet than anything on my accountants desktop.

Modern onboard computing is the ability to monitor the vehicle geometry (location, speed, rpm, braking etc) and driver behavior (hours of service, speed, cruise, lights, wipers etc) and bring the back office into the cab (driver messaging, pickup information, delivery information, signature capture, fuel tax, image scanning etc).  Seems simple enough, oh but wait, if it was simple all of us would be doing it.  You need to be aware that most trucking companies’ equipment committees are not tied to the backend accounting needs.  The push and pull is everywhere.  Traditionally if the rolling stock needed new toys, voila the money is/was found.  Over time this becomes a very complex dance between perceived value and benefit with real cost of support.

There are two main camps in this arena.  The first camp is all about driver and vehicle data.  The old world of tacho-graphs and vehicle data lead the way.  Our firm is squarely planted in this camp.  This is typically the camp of the early adopters.  Before backend systems had reliable means of real time connectivity the store and forward world of driver/vehicle data was one of batch reporting and driver coordinators who enforced compliance based on these reports.  This is not a bad camp to be in if it can move forward.  These types of solutions are usually very proprietary and closed and very good at collecting data.  Not so good at making information out of it though.

The next camp is the one that wants to bring the back office into the cab.  This is the where the late adopters find themselves.  The technology has matured and the industry as a whole is comfortable with baseline vehicle and driver information.  These systems typically are newer, open and standards based.  The support groups of trucking firms really enjoy these platforms for they are easier to support.  With anything new, there are issues to work out but all in all they are excellent approaches to solving real business information needs.

What about the driver.  In most worlds the driver is actually forgotten by the technology unless the firm forces driver feedback.  This feedback can be in the form of messaging, panic buttons for safety, audible alarms to enforce safety rules and give attention to immediate action.  Some of the technology is actually used to improve driver skills.  An example of this is lane departure software.  This type of software emulates the rubble strip on the sides of the highway so the driver is aware of any drifting.  It is very sophisticated in the math of determining if drifting is occurring and over time  the drivers have had positive feedback on its accuracy and then by their improved handling of the equipment.

So what is in it for the Driver?

1. Real time feedback on equipment issues.  For example spike break, lane departure, fatique
2. Panic button
3. Communications to and from dispatch in an non-obtrusive manner
4. Driver safety - Hi-jacking prevention and alarming
5. Driver pay based on performance
6. Reduced paperwork (driver logs, fuel tax, trip reports, pre and post trip inspections)

 So what is in it for the vehicle?

1. Effective and proactive monitoring of engine performance
2. Fuel usage
3. Safe handling
4. Extended vehicle life
5. Proactive and preventative maintenance procedures
6. Consistent, predictable usage pattern regardless of driver 

So what is in it for the organization?

1. Getting data as close to source as possible
2. Creating information from data
3. Increased customer service 
4. Customer transparency
5. Reduced labour to support operations
6. Real time tracking of fleet
7. Heightened security position
8. Better competitive edge
9. Ability to make decisions faster and with better context

 

Here are my answers to the questions.

Question 1: How has technology in the cab provided your organization a competitive edge?

Answer:  The technology deployed gives us near time information on business and operational performance.  Our customers have complete visibility into their freight.  Our bottom line is positively affected by lower costs, higher safety ratings, and predictable operational performance.  We are in effect proactive not reactive to the marketplace.

Question 2: Has technology enabled your company to operate in a safer environment?

Answer:  This was one of the biggest drivers for our onboard committee.  From driver fatigue, to driver safety, to safety compliance we are able to work with the driver to improve safety both for the equipment and more importantly for the driver.

 Question 3: What best practice has your company implemented to get driver buy in?

Answer:  We have a pretty proactive and progressive driver committee and driver coordinator team.  We run pilot projects and get driver feedback.  We adapt the technology to fit the needs of the driver and the organization.  We are always very aware of providing real time feedback to the drivers.  This can be achieved through visual or audio cues in the cab.

Question 4: How do you see government regulations affecting your business?

Answer:  We see the government finally taking the advice of our associations to adapt the technology in a supportive way.  To be more accepting of digital information.  We also see that there is still a lot of education required on the parts of government to understand what the industry is able to provide.  One of the challenges facing some of the early adopters and proactive firms is information overload to the government.  Some of the late adopters are just providing baseline information and not giving context to the information.  This is leading to a dual standard on the road.  Governments should align reporting requirements with reporting templates and guidelines. 

Question 5: How does the government inter-play affect your technology buying decision?

Answer:  More attention is being applied to things like driver logs and this is a good thing.  The late adopters have an advantage here where some of the early adopters are struggling with technology architecture issues.

Question 6:  How is technology enabling your organization to take advantage of fuel saving best practices?

Answer:  Along with safety this is the largest benefit achieved by onboard computing.  Elements such as idle time, progressive shifting, speed, excessive acceleration and rapid deceleration.  All of these elements have given real saving to fuel and repair costs.

In Conclusion

Onboard computing is maturing to a state that not having the technology is no longer an option.  We are at the stage that just having the technology is not good enough.  The firm must be able to leverage the data to make useful information with it.  Otherwise it becomes an exercise in diminishing returns.