PLM, Cloud, SaaS and Software Upgrades

I’m continuing to discuss various aspects of PLM on the Cloud and SaaS. The issue I wanted to discuss today is related to the software upgrades. This topic is considered as a painful in the enterprise software and requires additional significant investment and effort. Let think what SaaS/Cloud can change potentially? In my view, there are two separate aspects we need to analyze: technological and business.

Technology of SaaS/Cloud
From the technological standpoint, moving to cloud or SaaS solution still doesn’t mean change in comparison to what we have today when we run software on premises. If your PLM solution, for example, will be delivered on dedicated servers hosted in data centers, you are not going to experience any change in comparison to the web solution deployed in the orgnization. These servers will run the specific version of PLM software installed on them and you (or your provider) will need to care to upgrade them in the same way you are doing it today. However, if PLM solution will be delivered in the multi-tenant architecture, and you’ll share instances of the servers with other customers, you have a potential to experience non-stop migration between software versions deployed by vendors. Even in this case, the question of how it will happen to be heavily dependent on details of the system implementation, data modeling aspects and many others. So, in my view, the devil of SaaS/Cloud technology is in details. You’ll need to watch it closely and understand what can be supported by your software provider. Certainly, some of the software aspects related to data modeling will still require “upgrade point” anyway.

Cloud / SaaS as a business model
This is another aspect that I’d not connect directly to the technology of an upgrade. The most widely accepted situation when moving to SaaS model is to get software licenses paid by a subscription fee and not by license fees. These change of the business model will likely move your software expenses to a monthly/yearly payment. I had chance to read some interesting observation about that business from Burton Group blog yesterday. You can find it here. (http://ccsblog.burtongroup.com/collaboration_and_content/2010/02/a-.html). However, even this subscription looks like a big change, in practice, lots of SaaS contracts that promoted actively as monthly payments, will be considered as 1-2 years contracts. So, forklift event, Burton’s blog is writing about, will be created by vendor.

What is my conclusion today? Marketing is doing a great job around SaaS and cloud software. However, not all marketing presentations are translated into practical realities. It can stay in the power point slides. Even if SaaS/Cloud computing is providing an option to be different from software on premises, you need to watch details of a specific solution you are going to move to.

These are just my thoughts… What is your opinion on that?

Best, Oleg

Share

About these ads

27 Responses to PLM, Cloud, SaaS and Software Upgrades

  1. Herve says:

    Very good point my dear Oleg, the devil is in the details. Software vendors like Apple know very well how to deal with such migration issues: after all, the same operating system works for 3 generation of iPhone hardware, and migrating from 2.x to 3.x did not destabilize my phone.

    In the PLM realm, is any software vendor able to change its software version without a huge impact on data? I doubt so. The software maturity – or the data complexity – prevents so peacefull migrations. Maybe you’re right and the cloud will bring some good perspectives… So far, I guess Salesforce is the only one to be able to do it the easy way…

  2. Guy Alroy says:

    Dear Oleg, as far as the upgrade path, that goes back to the profile of the solution.

    From a user experience point of view, in a SaaS delivery aiming at SMBs, the vendor must make the upgrade path 100% transparent to the customer. I can’t see it working if not.

    From a technical point of view, a multi-tenant environment (both database level, application level etc.) enforces most typical upgrades to take place at a given time (usually at non working hours), and is not coordinated (but notified) with the end users. i.e., the next morning, when the user comes to work, the thing must be functioning…

    I’ve been using NetSuite for about 5 years now, from my perspective, the system may have gone through several upgrades, none required any action from me, maybe setting a parameter or similar. That’s how it should work :-)

  3. MarcL says:

    @Oleg – great to see someone finally focusing on the bane of the enterprise software world…

    as everyone knows, upgrades can take 6-9+ months for major global PLM installations depending on the level of customization. whether SaaS or on-premise the real issue is in the system’s architecture.

    if the system relies on conventional methodology where object model is hard-coded, then if customer does customization they are ‘breaking’ the object model. Now, they’re land-locked. next time a release comes out its incompatible with the changes made. in many cases, a highly customized global deployment effectively requires a re-implementation and data migration.

    Now, if the architecture separates the object model from the code, then you can upgrade the code without impacting the customizations (no matter how much customizing is done). It’s basically an “upgrade in-place” with no data migration, no programming rewrites, etc. Makes it fast for the IT team and transparent to the user (they probably don’t even know it happened).

    With this type of architectural approach the deployment scenario doesn’t matter: on-site, in Cloud, or hybrid deployment (i.e. vault behind firewall in on-site datacenter and application servers in cloud).

    This architecture is how Aras is implemented… and what it means for companies running our PLM system is that for highly-customized global deployments at scale upgrades for a major release take 1/100th of the man hours required for Windchill, Teamcenter, or others.

    Details matter, enterprise system architecture dictates a company’s ability to both customize and upgrade, whether on-site or SaaS/cloud.

    What’s your take?

    MarcL
    http://www.aras.com

  4. Herve, Thanks for the comments! Agree, Apple solved migration issues. The complexity of CAD/PLM and other engineering apps is on another level and data is one of the destabilizing factors to support smooth upgrades and migrations. Salesforce is a good example to learn, You are right!… In addition, moving to the cloud can force vendors to simplify their solutions. Best, Oleg

  5. Guy, Thanks for comments! Yes, this is how it should work :)… However, this is a challenging requirement for the multi-tenant environment working 24×7. I’d be interested to learn NetSuite experience. I’m not familiar with this product from the technical standpoint to understand what they are doing different to make it possible. Best, Oleg

  6. Marc, Thanks for your insight. In my view, devil is in the details. Architecture and level of customization are important factors. The level of customization can impact an ability to support system upgrades and migrations. SaaS vs. not-SaaS doesn’t really matter. The overall perception, for the moment, is that SaaS systems are less affected but mostly because of lower level of customization available. Best, Oleg

  7. Why any one is thinking about SaaS or Clouds?

    costing cutting factors in terms of 1.Implementation 2. upgrading 3.maintenance rather than any technical advantage, Of course we all know just its another “New” technology marketed very aggressively
    does that really serve the purpose? I don’t think so, same cream pie sold in different package by a different vendor,Does it come cheap? I don’t think so, Always things things are sold on “no of users” base, and still it will be sold on the same criteria.

    I see that, these will be technological experiments,since developers concentrate on “Webinisation” of technologies

  8. Chandrajit says:

    Thanks for your thoughts. I think there are two points needs to be address for establishing SaaS/Cloud service in PLM space:
    - first one is the consumer’s felling of losing control over system/Design Data/Application, as it may be hosted & managed by outside service providers.
    - second will be concern about data security, as one application may be providing services to various other customers/competitors as well.
    I think we need to slightly switch focus to develop more secure & quickly upgradable PLM solutions along with not losing focus on functionality improvement.
    These are just my thoughts…..!

  9. Krishna, I think there are several reasons why SaaS/Cloud got attention. 1/Cost; 2/Collaboration; 3/Availability; 4/computational power. However, I agree with you, cloud / SaaS can bring disadvantages or, at least, will keep the same problems we have with traditional approaches today. This is my exact point. Moving to the cloud/SaaS will not automatically resolve all problems related to upgrades of software. Thanks for your comments! Best, Oleg

  10. Chandrajit, Completely agree. These two points – security and control over data are main concerns of customers when talking about SaaS/Cloud apps. Thanks for discussion! Best, Oleg

  11. Excellent blog and comments!
    My biggest concern with this post is that it focuses solely on PLM systems hosted on cloud. If you put together all the other systems that are necessary for a PLM system to function like CAD, PDM, ERP, Integrations/Mashups…you still have a nightmare (large project) to manage.

    Upgrading one element of the puzzle involves testing and qualification of all touch points…this cannot be alleviated unless the whole eco system is hosted and the host provides complete service.

    let me know your thoughts…

  12. Arun says:

    Oleg, your overall question is very relevant in the context of cloud based applications. Just because an application is hosted does not take away the pain of application upgrades for its users. Full benefits of an
    hosted application can be best realized if there is one and only one version of the hosted application running and accessible to ALL customers. For users of such a hosted application, all application
    upgrades happen without them having to do anything about it. Such hosted multi-tenant applications need to be designed from the ground up such that it meets this key operational requirement. Such a requirement does not necessarily have to be done at the expense of customizations, especially if customizations are a core business requirement. That said, supporting customizations in a hosted multi-tenant does put additional burden during the design of an application feature. From looking at the comments you have received so far, looks like there is consensus that the architecture determines ease of upgrade and level of customization. Speaking from experience of the past 10 years, of having designed, developed and supported this model, I can certainly say that that this model works. Works for users and works for service providers.

    Best,
    Arun
    Arena Solutions

  13. [...] MatrixOne version. I want to point out on the discussion about PLM software upgrades – PLM, Cloud, SaaS and Software Upgrades. My conclusion was simple – technology and architecture matter. If PLM data management [...]

  14. Swati, In my view, enterprise systems are complex. Some technological thoughts how to make it less inter-dependent may be a good move… Do you know somebody is thinking about it? Thanks for your comments! Best, Oleg

  15. Arun, Thanks for your comments! Yes, sounds like we are in agreement. However, my question if large PLM implementations can be designed and implemented using such experience you mentioned. As I understand Arena is focused on the small-to-medium type of customers… Best, Oleg

  16. If you are shifting the platform to the cloud (just shifting the infrastructure doesn’t help as some already noted), then what you’re really doing is shifting the upgrade headache to the vendor. In a traditional model, everyone does their own installs and customization, then everyone has their individual upgrade headaches. If the vendor is doing a mandatory upgrade of central code that affects all end user company instances, then all systems need to work after the upgrade, or they won’t stay in business long. This puts a very strong financial incentive on the vendor to architect the system and their interface for customizing the software well to avoid the upgrade issue. They could already do this for on premises installed software, but before it was the end user company’s headache if they didn’t do it well. This shift of responsibility is what will get us software better designed for future upgradeability.

  17. Scott, Thanks for your comment! I cannot agree more. The cloud architecture needs to be designed bottom up to support upgrades and multi-tenancy for users. The regular “upgrade hammer” can stop working for such an environment. Best, Oleg

  18. Arun says:

    yes, Arena primarily targets small to mid-size manufacturers. That said, we have a number of large manufacturing companies who use Arena not only for the core features we offer, but also the ease and affordability of getting started and using the system. The platform itself does not limit the size of the implementation that can be supported.

  19. Krishna Badarla says:

    Cloud,SaaS, visualized as a virtual machine concept to reduce costs rather than any technical advantage,Just the ownership of the implementation, maintenance and up gradation related responsibility is taken over by the S/W provider,Does this solve any primary issues the Client is facing? I feel no.
    I agree with Arun, on the complexity of the Enterprise, Even though the standard(best practices) remain same, the approaches to the processes by the clients make it more complex, and leaves the system as an open ended, widening the gaps between the processes.
    Swati, I agree with you, as I was saying, the host never provides a complete solution and it remains open ended in terms of processes and data management, there are multiple reasons, like investment, technology development,and the time factor, the opportunity cost is very high and time bound, by the time, a concept is developed and offered, it is seen the technology is out dated, and especially in S/w, technology change is almost immediate and the risk element is: a particular solution may not be liked or may find inadequate by a client,(ex: CAD application), and rejects the whole system, this is very risky.
    So the providers try to limit the system generalized so that it can be easily marketed.

    The security aspect, well, it needs a lot of attention,Already there are allegations of data theft,the SaaS,and clod providers need to do a lot of work to prove that it is safe in their hands.

    Another point is, what If the user(client) wants to switch over to another system for what ever reason? and doesn’t like this system any more?,As the data is with the existing provider. who will provide the service of migration? the new vendor with whom the client fell in love or the old vendor whom he is divorcing? in this hostile conditions is the data safe?

  20. [...] few days back, Oleg over at Daily PLM Think Tank did another good post titled PLM, Cloud, SaaS and Software Upgrades. It looks at several different software delivery trends (i.e. on-premise vs in-cloud) to really get [...]

  21. Arun, Thank you for your comment! My assumption that bigger customers will have a different set of expectation related to the tuning of the system and specific customizations. Then, questions of upgrades will come again. Does it make sense? Best, Oleg

  22. Krishna, Thanks for sharing your thoughts! However, I don’t see the cloud and hosting as the only “cost” factor. Cloud (as well as Internet) can facilitate better collaboration and provide some solutions that can be different from the functional standpoint. The implementation/deployment time is different. And yes, it comes with additional issues such as security, etc. Good discussion! Best, Oleg

  23. Arun says:

    Oleg, Good question. Arena data model can accommodate customizations for many common processes to meet the unique work-flow requirement of our customers. When we make enhancements, and other changes to functionality that affect these processes, all of these customizations are upgraded as well.
    Regards,
    Arun.

  24. Arun, thanks for this clarification! I think that customization and multi-tenancy is the most complicated case for SaaS/OnDemand software. I was interested to learn more how Salesforce is doing so. It seems to me, they have the best solution based on Force.com today. Do you have any experience with that? Best, Oleg

  25. cloud computing services,cloud computing hosting,clouding computing,cloud computing service,cloud computing platform,cloud computing server,cloud computing security,cloud computing solution,cloud computing technologies,cloud based solution,cloud serv…

    [...]PLM, Cloud, SaaS and Software Upgrades « Daily PLM Think Tank Blog[...]…

  26. wordpress blogging…

    [...]PLM, Cloud, SaaS and Software Upgrades « Daily PLM Think Tank Blog[...]…

  27. reformas de casas…

    [...]PLM, Cloud, SaaS and Software Upgrades « Daily PLM Think Tank Blog[...]…

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 237 other followers

%d bloggers like this: