The Cloud's Silver Lining for CAD22 Aug, 2017 By: Robert Green
CAD Manager Column: Private application clouds, rented virtual workstations, and remote server rentals are three cloud-enabled technologies that can make CAD managers’ lives easier instead of more difficult.
As you learn more about cloud technologies for CAD applications, you may start to focus on how difficult they are to evaluate and implement. It’s impossible to know exactly how new technologies will work out, so it pays to be skeptical. But if you let yourself consider only the negative aspects, you may miss the silver linings of the cloud.
But what are those bright spots, exactly? Are there any cloud-based technologies that CAD managers can look toward with optimism and enthusiasm? In this edition of the CAD Manager’s Newsletter, we’ll explore a few such possibilities that I’m seeing more frequently. Here goes.
Private Application Clouds
First, let’s define what private cloud means: A private cloud is a resource located somewhere in your company’s IT infrastructure; remote users can access it via a virtual private network (VPN). Think of it like a remote drive on your corporate wide-area network (WAN). The key point is that the resource lives inside your company’s IT environment, not in some cloud vendor’s server farm, and therefore it’s totally under your control.
A private application cloud, then, is simply a software application that resides on your company’s private cloud. Consider the following examples:
- A visualization workstation. Install a high-performance machine equipped with specialized graphics processors and visualization applications somewhere in your IT’s data center, then give your remote users access to it.
- Analysis tools. Share a structural engineering application or finite analysis tool running under Autodesk Revit or Dassault Systèmes SOLIDWORKS so that engineers around the company can share the software.
- Specialized video or animation tools. Share high-performance workstations with solid-state drives and super-high-performance graphics processors for video editing or animation tools.
In these cases, several advantages emerge:
- Lower software costs. By sharing specialized software among many users instead of buying a license for every potential user, you’ll save money.
- Hardware savings. Sharing high-end workstations outfitted with expensive graphics processors, rather than investing in a machine for every user, will also save you money.
- Less waiting. Since the software and all the data files will reside on the private application cloud, there are no lengthy upload and download wait times (as there are with many cloud architectures).
- No increased threat to data security. Since your data never leaves your server, hacking and data theft concerns are no different than with any other in-house IT process.
- Minimal learning curve, plus perpetual savings. Since this approach uses software you already use, there’s no need to learn new complex cloud applications. And for extra savings, you can continue with perpetual licenses rather than renting cloud software.
I’m seeing more and more implementations of private application clouds, and that’s no accident. The economic, speed, and security advantages of this approach are significant.
No matter how many workstations you have, it seems you could always use more of them, or at least some faster models! Well, with workstation rental, you could simply beef up your workstation portfolio for the short term. And with technology advances, it’s safe to assume that more and more cores and processors will become available as workstation rental becomes more common.
Admittedly, this is a new market, but is it really that far-fetched to rent a workstation and let the rental company deal with the maintenance and hassles? Cloud-based CPU farms make this scenario more and more feasible, as service providers configure high-powered workstations with popular software that you can use in the short term for reasonable fees.
The possible advantages include:
- No machine ownership. That’s the vendor’s problem.
- No IT headaches. No configuring graphics cards, network interfaces, etc. — all that work is done at the CPU farm.
- Conserving cash. Pay just for what you need, at a fraction of the price of ownership.
- Your workstation resources can be shared across offices, countries, and time zones. Your Phoenix office could run for eight hours, then hand off the resources to your team in Singapore, then the German branch can take a turn.
- Your data ownership is maintained. Simply upload your software/models and run your jobs, then download your results.
- Security risk is minimized. Your data is only on the cloud for as long as your job is running on the remote machine. After downloading, your data is back in your network, so any exposure you may have is greatly reduced.
- Speed and bandwidth issues are minimized. Since initial upload and downloads are the only time when these issues matter — rather than continually moving application data over an Internet connection — CPU rental is no more data-intensive than the current uploading/downloading we experience today.
While the cost of standard CAD workstations is very affordable for most users, the machines used for high-end analysis or visualization tasks are still very expensive. When renting these configurations is as easy as uploading and downloading some files, why would we suffer the drudgery of processing on old hardware, or spend a fortune on something we won’t use often?
Remote Server Rentals
For companies that have smaller branch offices, it can sometimes make sense to rent a remote server rather than use servers at the main office. Why? Consider the following advantages:
- Bandwidth is saved. When the branch offices use a rental server, they are not going through the main office IT infrastructure, so the traffic at the main office is reduced. And even when the main office IT Internet speed is low — or is offline altogether — the remote server will continue to function normally.
- Applications can be hosted. Since a real server is used, all business applications that are server-based can be deployed on the server as well. This may include CAD applications or project databases. The point is, you’ll have all the flexibility you would with any other type of server.
- Latency may be reduced. If a branch office in San Diego rents a server located in Los Angeles, the round-trip latency time to the rental server would be far less than the time for connecting to a corporate server that might be located in Chicago, for example. Lower latency and round-trip times make it far easier to utilize private application clouds that could be housed on rental servers.
Of course, your company IT staff will still have to administer the server, but given the other advantages, rental servers can make a lot of sense.
Obviously, the cloud remains a work in progress, with many vexing issues to resolve — including security, latency, and bandwidth problems. Certain applications, however, such as private application clouds and workstation and server rentals, are starting to become more mainstream. My hope is you can think about how these examples might help you get more for your software investment, better share resources, and bring more hardware horsepower to bear in your departments.
Please e-mail me your thoughts at firstname.lastname@example.org, or swing by my Facebook group “CAD Managers Unite!” and leave your feedback there. You may even see your name in the next newsletter. Until next time.