How cloud CAD can avoid “double PDM tax”

March 5, 2015

cloud-cad-API

My yesterday post – Will cloud CAD inherit data interoperability problem? raised few interesting discussion about cloud data management in PDM/PLM implementations. How cloud CAD/PDM will make our life simpler? In my view, the most important part is to exclude files from data management chain. By doing that, new cloud based CAD systems are able to make data flow much easier. Existing check-in/out behavior will become redundant in cloud systems, cloud applications can save data instantaneously and redundantly and will allow you to restore to any point of changes. I also hope cloud CAD systems will have lock function in case you want to prevent somebody else from changing your design. The same mechanism will also provide a way to branch design options much easily by leveraging direct data access to all design data stored in the cloud databases. This is my dream scenario.

However, data interoperability of new CAD/PDM bundles seems to be a potential point of failure. And it is can slow down adoption of cloud CAD systems in environments that require integration with existing desktop CAD, PDM and PLM systems. The following Engineering.com article can give you some context to the problem -Dassault or Siemens PLM? The Contrasting Paths of Jaguar Land Rover and Volvo Cars. It speaks about challenges of large manufacturing companies related to usage of CATIA V6 / ENOVIA PDM. Here are few passages that caught my attention:

Volvo invested in Siemens PLM solution Teamcenter as a backbone, and kept CATIA V5. However, the automaker is reluctant to switch to CATIA V6 and the 3DEXPERIENCE/Enovia V6 platform. "We will not use the V6 version if it requires double PDM installations", says VCC’s Andreas Westholm, IT Director – Geely Liaison.

Volvo will not use CATIA V6 if it requires a second PDM implementationAll CATIA files are managed in Teamcenter. Since Volvo does not have any plans at this time to migrate to CATIA V6, they don’t need Dassault’s Enovia PDM as an intermediate step in the data management.

”It is not possible to work effectively with two PDM systems”, asserts the Volvo IT-director. ”And we will not use CATIA V6 if it requires double PDM installations. However, we will bring in a new V5-V6 release that facilitates the import of V6 information”.

Potentially, any cloud CAD (with embedded PDM functionality) can create a situation similar to CATIA V6, which is a problem. Engineering and manufacturing companies have very slow process of new software adoption. So, to be successful, cloud CAD systems will have to co-exist and be used alongside with existing desktop CAD systems. What is even more important, new cloud CAD systems will have to be integrated with existing PLM products to become part of product development processes. How to prevent future cloud CAD systems from a problem described by Volvo? How to avoid future "double PDM tax" on cloud CAD systems?

I think the answer is in a new cloud system architecture. It reminded me one of my old posts – Why PLM needs to learn Web APIs? A potential solution to the double PDM integration problem is future cloud CAD platforms, web APIs and data openness. Think about the way most of modern web platforms are consuming data. Seamless data streaming, avoiding local temp file storage and standard REST-based API is allowing us to create better integration between web systems. This is a way new cloud CAD solutions can be seamlessly integrated into existing PLM solutions and eliminate "double PDM tax".

What is my conclusion? Future of cloud CAD/PDM bundles is promising and can provide many advantages to users – transparent data management, ease of revision management and collaboration. However, it is very important to think how new cloud solutions will be integrated with existing PLM platforms. Openness and web-based APIs are two most critical elements to support integration and adoption of new systems. Just my thoughts…

Best, Oleg


Will cloud CAD inherit data interoperability problem?

March 3, 2015

cloud-cad-pdm-interop

Cloud and CAD are probably getting to the point where it starts become a real thing. Autodesk Fusion360, Onshape, SolidWork Industrial design. It is likely to absorb some PDM functionality to make collaboration, branching, revisions and other data management tasks easier. Cloud CAD means no files, so engineers have nothing to mess with… Life is getting more Googley if you read John McEleney Onshape blog.

However, here is the thing… What if (for some crazy reason, which is easy to imagine when you deal with engineers :)), customer will decide to do a work with two cloud CAD systems? It is not unusual to see multiple desktop CAD systems in engineering organizations, so why cloud CAD will be any different.

In my yesterday blog – Cloud CAD infrastructure is getting more PDM-ish, I draw the picture of cloud CAD/PDM bundle helping us to collaborate and manage revisions. Now how two cloud CAD systems will work together? I’ve been trying to bring my cloud imagination and thought about Google Docs and Office 365 services co-existence. Actually, it is not very nice story- I can easy get my files distributed between my Dropbox, Google Drive and OneDrive accounts. So, what if my parts will be stored on Google Drive and Assembly on Dropbox? Not sure I will like it…

Similar problem in PLM world created many debates and issues. Do you remember Dassault CATIA V6 story, which required ENOVIA backend to run it? It made few customers unhappy when they discovered that they need to run two PDM/PLM systems. I can see some similarity with multiple CAD/PDM cloud bundles co-existence and interoperability.

What is my conclusion? How engineers will collaborate using multiple CAD cloud software? Cloud technology is great, but it looks like cannot magically resolve some old fundamental problems of multiple systems, collaboration and interoperability. I wish cloud CAD / PDM vendors will think about it upfront before customers will find themselves in the middle of messy CAD / import/export/migrate data scenarios. Just my thoughts…

Best, Oleg


Cloud CAD infrastructure is getting more PDM-ish

March 2, 2015

fusion360-cloud-pdm-cad-branches

PDM was long time a step child for many CAD systems. To deal with CAD files, their relationships and dependencies including multiple revisions of document was complex and painful. So, many customers just gave up and stored files on shared drives. It was okay until the our life was disrupted by a new way to get work done – online, connected, collaborative.

The initial intent of collaborative cloud systems was to solve the problem of collaboration and data sharing. The idea to provide a value on top of existing CAD desktop file systems was interested. So, specialized cloud CAD file sharing and collaboration systems got focus and attention.

At the same time, CAD vendors got to think about longer term solutions. CATIA was one of the first systems that announced the disconnect from a traditional file systems. I posted about that few years ago – The future of CAD without files.

These days, the race towards cloud CAD is accelerating development of data management and CAD collaboration technologies for the cloud. Few weeks ago, I shared some of my thoughts about importance of PDM technologies for cloud CAD. Cloud CAD vendors are clear about their intent to make PDM part of their core product technology.

My attention caught Autodesk Fusion360 article – Fusion System Architecture Changes Coming in the Next Release – Why and What. Read it, because it contains some very interesting pieces of information about how files are going to disappear into future cloud infrastructure. Here is the passage I captured:

With the coming release of Fusion, we will be introducing significant changes to Fusion Cloud Service architecture that lays a strong foundation on which we can build an environment that is rich in WIP DM. As part of this change, we are introducing the Autodesk Work In Progress Cloud Service which is designed to model and manage complex relationships that are associated with a design. The service is highly scalable, highly available and optimized for performance. Another important change in the February release is significant improvements to the Fusion Object Storage Service. Taken together, these changes will result in immediate benefits in the way of performance gains and high reliability in the Fusion upload and download data pipeline, and allow the Fusion team to deliver rich DM workflows in subsequent releases.

Another article from Fusion360 blog brings an excellent explanation what these date services mean for end user. These are functions that belonging to PDM system in a traditional file based CAD / PDM setup.

…it solves so many common design problems that we’ve heard from the community, both in Fusion 360 and other programs, and improves workflows for both teams and single designers. Branching and merging lets you easily:Work in parallel with other members of your team. Explore changes or alternatives to a project and keep changes that make sense while leaving behind changes that don’t. Understand how your project evolved over time and what decisions were made (and why). Restore or reuse any design(s) in your project from any point in your project. Use any point in your project as a starting point for a different project.

What is my conclusion? It is hard to bring a value of cloud design collaboration without re-thinking the way CAD-PDM bundle is operating. For new type of cloud CAD systems it means to embed core PDM collaborative function and make it part of CAD system. It sounds like a very exciting time- many collaboration and data management problems are going to be solved by that. However, here is a question. What will happen when two cloud CAD systems will have to collaborate together? Looks like a topic for another blog. Just my thoughts…

Best, Oleg


What will change existing PLM paradigms?

February 26, 2015

It is not uncommon to hear about “changing paradigms” in different domains these days. We are watching Netflix and disconnecting cable TVs, using Uber instead of driving our own cars. Yesterday at PI Congress, I saw the following slide demonstrating examples of digital disruption in different industry domains.

digital-disruption-pressure

Which obviously made me think about disruption in PLM. This domain has some characteristics that make it hard to disrupt. 1 / It is dominated by a small number of very well established vendors. 2/ The barrier to entry the space is high in terms of expertise and completeness of the solution. 3/ Decision lifecycle for customers to buy a software is long and the usage lifecycle is even longer. Companies can use software for 10-15 years because of product lifecycle (eg. aero-planes). As a result of that, one of the main drivers to change PLM system is in fact because existing PLM software will no longer developed or supported by PLM vendor.

For the last decade, we’ve seen very few example of starting a fresh new paradigm in PLM system. Aras Corp came with enterprise open source Aras Innovator. It was a cool idea – think about “Linux of PLM”. It would be interesting to see how much focus Aras will put in their open source in the future.

Another fresh start was Autodesk PLM360, which introduced “cloud PLM alternative”. Even ideas of “cloud” or “hosting” aren’t new and some vendors in PLM space did it before, entrance of such a big vendor like Autodesk in this domain made a change in the industry. 3 years later, we can see all PLM vendors have “cloud” in their portfolios.

There is one thing that didn’t change in PLM and this is very painful thing. You cannot just install and start using PLM like email. In the world of PLM it called “implementation”. You need to figure out how PLM products will help to organization to use it for their product development processes. And this is all about people. Technologies are easy, but people are really hard. Therefore, in my view, PLM got stuck with people. The current paradigm assumes PLM implementation as a core fundamental part of everything. It slows down adoption and requires extensive resources and effort from organization. How to change that?

Have you heard about DevOps? If not, I recommend you to put aside whatever you do and close this educational gap. It is well known in software development and it is essentially a combination of two terms – “development” and “operations”. It became popular and it is a result of massive introduction of new software development practices combined with cloud operations. Few months ago, I mentioned devops in my post – Why to ask your cloud PLM vendor about devops and kubernetes? Business insider article Today’s IT department is in fight for its life helped me to bring my thoughts to clarity. Here is my favorite passage.

Devops is all about how do things faster,” Red Hat CEO Jim Whitehurst tells Business Insider. It’s the IT department’s version of Facebook’s famous mantra “go fast and break stuff.” IT departments say they had better figure out how to be faster, cheaper, and better. If they don’t, the company’s employees will no longer depend on them. They bring their own PCs, tablets and phones to work and they buy whatever cloud services they want to do their jobs. And the CIO will find his budget increasingly shifted to other manager’s pockets.

“Like the manufacturers were in the 1970s and 1980s were fighting for their lives, today’s IT departments are going to fight for their survival,” Whitehurst says. Traditional IT departments are slow and methodical. Rule no. 1 was to never bring the systems down. They would take months, even years, to roll out new new software, testing everything carefully, often spending millions in the process. Devops eliminates that. Instead, IT departments tear their projects apart into teeny components that can be implemented in tiny changes every day.

The last phrase is a key one. How to tear projects apart into teeny components to be implemented in tiny changes. It made me think about existing PLM implementation paradigm. It heavily relies on long planning cycle and business department alignment. Once this planning made, implementation takes long time and put ROI in absolutely wrong place from what organizations are demanding it to be.

So, how PLM can adopt new way to do things? It requires 3 main changes – 1/ To change state of mind. Don’t think “one big implementation”. Opposite to that, think about small steps that will make business better, faster, efficient. 2/ To bring new PLM biz development tools that can help organizations to plan into small steps. 3/ To make PLM platform capable to function in Devops mode. It requires new type of data modeling, deployment and monitoring tools.

More to come, but I think, Devops ideas can inspire and educate PLM developers to think differently. How to develop PLM practices in a different way. How to bring a new feature in a day and how to test changes for the next hour. These are questions PLM business consulting, developers and business consulting should ask about.

how-to-change-plm-paradigm-with-devops

What is my conclusion? Changing paradigms is hard. For many years, PLM industry fundamental paradigm was to relies on implementation as adoption process of PLM technologies. It started from selling PLM toolkits that required long implementation. PLM vendors tried (still do) out of the box approach, which mostly ended up as a good marketing to demonstrated capabilities of PLM technologies, but required implementation anyway. Cloud approach cut the need for expensive IT involvement, but still requires implementation process. PLM industry needs to find a way to make PLM implementation simpler and easier, so people will stop thinking about PLM implementations as a mess. Just my thoughts…

Best, Oleg


New enterprise software reality for PLM vendors

February 21, 2015

enterprise-software-new-reality

I’m planning to attend PI Congress event in Dusseldorf next week. As part of the preparation, I’m taking a look back on past events. Back in 2012, I was presented at PLM Innovation 2012 event in Munich about PLM and consumerization. You can navigate and see my presentation. One of the topics I discussed was related to enterprise software and future impact on enterprise IT. The changes were coming to enterprise IT and we can see it clearly now. Business insider article Today’s IT department is in fight for its life can help you to understand the significance of changes.

enterprise-IT-on-fire

Changes in enterprise IT made me think again about what changed in enterprise software. I want to come with top three significant changes comparing to 2012 – Distributed, Connected, Cloud.

what-changed-in-enterprise-software

1- Distributed. Nothing happens these days in a single place. Regardless on what type of company and industry sector are you working, you have to deal with the situation when you customers, partners, suppliers, engineers, manufacturing facilities are located everywhere. This is a new reality and it is getting reflected into enterprise software.

2- Connected. Another thing. People need answer now, at the right moment. People are expecting a different speed from business and enterprise software must provide an answer. Social networks and web disrupted the way enterprises are operating nowadays. Everyone is staying connected.

3- Cloud. Enterprise moved from discussion about cloud into practical implementation of cloud software. It is not about why to implement cloud. It is about how to do it right. Enterprise software vendors and their customers are looking how to find a right solution for today and tomorrow.

What is my conclusion? The reality of today is distributed enterprise software. Essentially, there is no single location for any organization. If you developing PLM software today, take the distribution factor as the most fundamental requirement. Second is communication – the information is flowing between people at high speed and frequency. The right information is in chat, messenger, shared content. How to pickup this information and make it connected to people decisions is a critical aspect. The last one is cloud. There are lot of confusion about what is cloud and how it can be done for organization. But, every organization is going to discover how cloud is impacting what enterprise IT is doing. No escapes here. Just my thoughts…

Best, Oleg

Image courtesy of KROMKRATHOG at FreeDigitalPhotos.net


Why PLM got stuck with people?

February 11, 2015

plm-got-stuck-with-people

I want to step back and look on PLM development trends for the last 4-5 years. Cloud PLM was one of the most visible one. It came as a result of enterprise software consumerization – the trend that made lot of mature web and cloud technologies available for enterprise organization. The adoption of cloud technologies by PLM vendors is accelerating. It became clear that none of large PLM vendors is ready to leave such competitive advantage such as “cloud” to others.

Regardless on technological and product maturity, all PLM vendors today are providing some sort of “cloud PLM” today. Also, there are running debates about what is “real cloud” and what is “pseudo-cloud”. I can see some valid points in these discussions, especially when it comes to availability, browser support, upgrades and the most important element – cost. I don’t want to judge PLM vendors in their run for cloud tech for any cost. However, after dust will settle in cloud adoption process, vendor’s cost and infrastructure utilization will become of the most important competitive elements. They challenge for vendors will be to provide the most “economically reasonable” PLM solution.

I was thinking about first round of cloud PLM adoption. Let’s call it Cloud PLM 1.0 for this conversation. It made lot of good things – customers now can start implementing and using PLM without high upfront cost. It allows global access to information regardless on location, it removes customer challenges related to upgrades because vendor is essentially responsible for a solution and sell it to you as a service.

Do you think we live in PLM nirvana now? Not so fast… There is something cloud PLM cannot do for you. And it comes to the point of how to get PLM implemented for your company. Implementation of business processes is an interesting aspect of adopting PLM. In practice it means to define data structures and business processes. This process is essentially the same for both cloud and non-cloud PLM systems. Cloud based systems might have advantages related to the way system can be administrated. But, this is not something that impossible for non-cloud solution. It is all about people, processes and organizational changes. So, the ugly truth is that cloud PLM won’t reduce your need of implementation services. In case of on-premise PLM, implementation will be done on site and collaborate with IT – installing, configuring and debugging customized software. In case of cloud PLM, you will need to work with cloud PLM vendor or hosting provider.

Earlier today, I was reading Aras PLM whitepaper called – Is your PLM initiative stuck? Navigate to the following link to download a whitepaper. It speaks about top 10 typical ways your PLM initiative get “stuck”. There is one thing that struck me the most when I was reading this white paper. Majority of issues are clearly related to people in an organization that adopting and implementing PLM. Software vendors, service providers and consulting cannot solve this problem. You need to solve it in the interaction with customer. Here are some examples – poor planning, taking decision to automate everything because you can, the decision to take one giant implementation step, no long term vision, too long consulting stage with no “go live” moment, decision to implement fragmented “point” solution, etc.

What is my conclusion? People is the most complicated part in PLM paradigm we have today. Regardless on a specific PLM system, PLM implementation process is heavily dependent on people and sensitive to many things – ability of organization to understand their problems, internal organizational politics, building consensus within organization, aligning requirements between departments, divisions, suppliers and contractors. All together it creates a potential for implementation failure. How to solve this problem? Would it be a magic technology that will create a complete new paradigm for PLM? Maybe it will be a well defined blueprint for implementation that company can easy adopt? To remove PLM sensitivity to people factor is an interesting challenge PLM vendors will have to deal sooner than later. Just my thoughts…

Best, Oleg

Image courtesy of iosphere at FreeDigitalPhotos.net


Cloud CAD will have to solve PDM problems at first place

February 10, 2015

ds-solidworks-cloud

The race towards CAD in the cloud is getting more interesting every day. I’ve been watching SOLIDWORKS World 2015 live streaming this morning. Overall SOLIDWORKS show was very impressive, as usual. I look forward to keep watching it following days.

However, what caught my special attention today is a presentation of a new cloud product – SOLIDWORKS Industrial Design. Couple of interesting facts about new product – works on top of Dassault 3DEXPERIENCE platform (in many aspects, think about it as ENOVIA V6); focus on free design with no traditional CAD constraints; leveraging cloud approach for social connection between users and collaboration. And… it is complimentary for SolidWorks users. Demo of product presented few scenarios in which design flow went between new cloud product and old SolidWorks connected by 3DEXPERIENCE platform. Sounds like a big deal.

The story about new SOLIDWORKS/3DEXPERIENCE product took me back to my comparison of Onshape and Autodesk Fusion360 visions few days ago – Carl Bass and Jon Hirschtick are in agreement about future of CAD. In my view, changes in design world towards distributed teams and ability to work effortlessly on any device without installation and manual upgrades are two main driving factors behind new cloud solutions. The story about Autodesk Fusion360, Onshape and SOLIDWORKS made me think about interesting priorities all creators of cloud CAD are thinking about. It was well articulated during SolidWorks World 2015 first day keynote – supporting any device, information is up to date all the time, users are connected (see picture above).

It makes a perfect sense to me, since it is a great reflection of modern cloud paradigm you can see well developed in products such as Google Apps, Office 365 and others. However, this is a place where complexity of CAD data requires from cloud products to be more sophisticated. Traditionally, CAD project is combined of multiple files – assemblies, parts, drawings, etc. As you start making changes you very quickly end up with a complexity of many-to-many relationships between different versions of parts, assemblies and drawings. Move it to the cloud – the complexity won’t disapear. Therefore, you can see both Autodesk Fusion360 and SolidWorks Industrial Design are trying to solve. I had no chance to see Onshape product yet, but my hunch Onshape will try to solve this problem too.

Autodesk Fusion360: revision control, branching and collaboration

autodesk-fusion360-rev-branches

carl-bass-fusion-360-au2014-2

SolidWorks Industrial design: 3DEXPERIENCE collaboration, branches, revision merging

solidworks-industrial-design-1

solidworks-industrial-design-2

What is my conclusion? In a traditional CAD world, the problem of file revisions, collaboration and data control was part of PDM solution. Historically, CAD vendors were reluctant to solve PDM problems unless it became absolutely necessarily. PDM was complex, required services, special pre-sale process, etc. However, cloud is creating a new demand and constraints for new CAD in the cloud paradigm. With the absence of file system exposed to end user, cloud CAD system will have to solve a PDM problem first. Just my thoughts…

Best, Oleg


Follow

Get every new post delivered to your Inbox.

Join 273 other followers