Revit Cloud Worksharing Forum
Welcome to Autodesk’s Revit Cloud Worksharing Forums. Share your knowledge, ask questions, and explore popular Revit Cloud Worksharing topics.
cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Is C4R still Really a good idea?

13 REPLIES 13
Reply
Message 1 of 14
dtpeter2901
1489 Views, 13 Replies

Is C4R still Really a good idea?

Just a general question/rant that I feel needs to be asked of someone at Autodesk in hopes of an actual Answer?

 

I'll start by saying when it works, I do like the product.  

For the first Year we used it successfully on one project as a means to test out the services.  Other than the growing pains of getting up and running, there was only one "glitch" in the first year or so we used the product (2015-2016).  The service worked well.  We never ran into issues with models getting corrupt.  Heck we even sold it to a client as a means of us doing a better job and won the project because of it.  We started with just 8 Seats.  I think we now have about 90.  It's a good solution when it works.

 

However, these last few weeks have been very painful.  There was also a time (I think about April) when for about 2 full days I had several users that were denied access to projects saying they didn't have a license, when in fact they did.  Last week I created a new project on A360 (BIM 360 or what ever you want to call it now) and it let me do so.  Right after I created it (5 min) I tried to invite our other BIM staff to the project and it said I didn't have access to it.  Later it did.  This has been a very painful 2 weeks.  Models randomly getting corrupted and rolling back the project doesn't seem to work.  Which make me ask the question, if it wasn't corrupt when a person at 8pm saved to central and was the last person in it, and it's "corrupted" the next morning, how's that happen.  Then you try to roll it back 6 versions (maybe 4-6 hrs) and it's still says it's corrupt, can someone explain that?  I'm able to use a local file that was saved out and that's got no problem.  These things usually have been happening when A360 says it's has a "Glitch" or someone can't publish or something that sounds rather benign like that.  It really starting to make me ask the question if I've been wasting my employees time by rolling these projects back when really it's been an issue with the data transfer of the A360 sever and nothing that we've done.

I'm also wondering if Autodesk didn't plan for the scale of use.  I'd also like to know how Autodesk goes about testing these changes (if they even do) before pushing something live.  

 

All of this makes me ask the question if it's really all worth paying $800 per seat per year so I can roll models back and waste my time writing these silly rants about why I need to pay for a product that seems to be inhibiting my productivity.  Reminds me of the Onion Sony commercial.

 

Yours Truely

One Really Annoyed BIM Guy.

13 REPLIES 13
Message 2 of 14

Thanks for taking the time to make this post.  I don't consider ths a rant so much as a statement of your frustration about your team's recent experiences.  I see two major points of conversation to have in response:

 

1.  Cloud Worksharing Availability & Reliability

The previous 6 weeks of service for C4R have been the most challenging in the service's lifespan, and your sentiment here reflects that of other customers we've heard in that timeframe.  As we've been on record before previously, we understand that confidence and trust are some of the primary factors for companies like yours that make subscription decisions, and the teams that put those subscriptions to use on projects.  This is not a new point of understanding for us, and thus has been a primary influencer in the investment decisions we make around our product teams.  There are many workflow and feature enhancements around C4R that have not been executed as a result of tradeoffs for that focus, and we continue to work to ensure that Cloud Worksharing services delivered by C4R are the best, most reliable, option relative to the other available options on the market.

 

Specific to the recent outages, if it's any consolation (I'm guessing it's not), it's not been an artifact of load or scalability, but primarily resiliency of Cloud Worksharing services in the face of outages of dependent services.  The past 3 months of investment have been almost exclusively focused around that area, and we hope to deliver that value to subscribers like yourself shortly.  Reglardless of the cause though, what matters is trust and confidence, and like I said back in July 2015, it's up to Autodesk to engender that trust and confidence that goes along with the game-changing collaboration that C4R delivers.

 

2.  Model Corruption

To my knowledge, I know of no situation in today's latest C4R code where model corruption happens as a result of cloud worksharing transactions.  Model corruption, like service availability and reliability, are key to trust and confidence, and we take any such instances extremely seriously across the Revit team.  As such, we'd very much like to investigate the experience you had, as it's contrary to the investment and testing that we do in this area.  Please send me a private message with your support case ID and I will make sure we take a look at what happened.  Additionally, it's a good idea to ensure your teams are running the latest Revit & C4R addon versions for the respective major version you are running on your projects.  I know we had one defect back in February - since fixed - that was resulting in a low incidence or corruption, and perhaps that might be the root cause of your particular issue.

 

We look forward to hearing back from you, and re-building the trust and confidence that's been lost for you recently.

 

Sincerely,

Kyle



Kyle Bernhardt
Director
Building Design Strategy
Autodesk, Inc.

Message 3 of 14

I just moved a number of teams over to C4R and they are now sitting around with nothing to do. While I get these silly images posted about how many more minutes until the site will be back up. 

 

If C4R is not ready for prime time, then let folks that depend on publishing their work for a living know that. But all Autodesk road maps lead to the cloud. It looks like that road might lead to a dead end!

 

Hopefully Autodesk will start offering compensation for these outages, as they directly impact a businesses bottom line. Which is the point of using the cloud in the first place. 

Message 4 of 14


@mike.hnastchenko42DZM wrote:

I just moved a number of teams over to C4R and they are now sitting around with nothing to do. While I get these silly images posted about how many more minutes until the site will be back up. 

 

If C4R is not ready for prime time, then let folks that depend on publishing their work for a living know that. But all Autodesk road maps lead to the cloud. It looks like that road might lead to a dead end!

 

Hopefully Autodesk will start offering compensation for these outages, as they directly impact a businesses bottom line. Which is the point of using the cloud in the first place. 


Cloud worksharing operations continue to operate normally, but you are right that the BIM 360 Team product is currently unavailable.  If you need to get the RVT files for sharing to other stakeholders, you can always open -> Save As, which creates a detached copy you can exchange.  If you need the web viewing or download capabilities, then I can't provide an interim solution until service is restored.  I know these are not great answers, just trying to go our best to help you out with other short-term options.

 

-Kyle



Kyle Bernhardt
Director
Building Design Strategy
Autodesk, Inc.

Message 5 of 14

Users that were not logged into C4R before the outage, cannot access their files from Revit. The A360 icon say they are not assigned to any projects. So the users that are currently in a project I told them not to exit out.

Mike

Mike Hnastchenko, AIA
Director, Strategic Solutions
M +1 612 396 0185
wd
wdpartners.com
What's Next from WD!



This transmission may contain information that is proprietary, confidential, and/or privileged and is intended only for the person or entity to which it is addressed. If you are not the intended recipient, you are hereby notified that any review, disclosure, copying, distribution, or use of the information contained herein is STRICTLY PROHIBITED. If you received this transmission in error, do not read it. Please immediately contact the sender and destroy the material in its entirety, whether in electronic or hard copy format, including attachments, and delete the material from any computer.
Message 6 of 14

Mike,

Our analytics data doesn't indicate any widespread incidence of such failures at the moment, so we need to investigate what you're seeing.  Can you post a separate thread and post a journal from a team member that's being denied access?  We can take a look there.

 

-Kyle



Kyle Bernhardt
Director
Building Design Strategy
Autodesk, Inc.

Message 7 of 14

Looks like the team members can now get into C4R.

A360 is still down so I can't add anyone new to the projects.


Mike Hnastchenko, AIA Director, Strategic Solutions






Mike Hnastchenko, AIA
Director, Strategic Solutions
M +1 612 396 0185
wd
wdpartners.com
What's Next from WD!



This transmission may contain information that is proprietary, confidential, and/or privileged and is intended only for the person or entity to which it is addressed. If you are not the intended recipient, you are hereby notified that any review, disclosure, copying, distribution, or use of the information contained herein is STRICTLY PROHIBITED. If you received this transmission in error, do not read it. Please immediately contact the sender and destroy the material in its entirety, whether in electronic or hard copy format, including attachments, and delete the material from any computer.
Message 8 of 14

As of ~4am EST BIM 360 Team is back up and operational.  C4R teams can now administer Projects, upload/download, view & markup models.  We are seeing a continue degradation of publishing to BIM 360 Team, but that is not a full outage at this point.  Our teams are investigating right now.

 

-Kyle



Kyle Bernhardt
Director
Building Design Strategy
Autodesk, Inc.

Message 9 of 14

Hi Kyle,

C4R as you know is down again.

And that's all the info we get.

I've got people in our model currently and we're wondering what's the best option for them.

Continue to work and not save or just hang out.

I understand sometime Adesk knows what the issues are and tries to list that.  But maybe offering a suggestion or just more information might help us user folk become less frustrated with not know what to do.  

Thanks in advance.

 

Message 10 of 14
chubbard
in reply to: KyleB_Autodesk


@KyleB_Autodesk wrote:

Thanks for taking the time to make this post.  I don't consider ths a rant so much as a statement of your frustration about your team's recent experiences.  I see two major points of conversation to have in response:

 

1.  Cloud Worksharing Availability & Reliability

The previous 6 weeks of service for C4R have been the most challenging in the service's lifespan, and your sentiment here reflects that of other customers we've heard in that timeframe.  As we've been on record before previously, we understand that confidence and trust are some of the primary factors for companies like yours that make subscription decisions, and the teams that put those subscriptions to use on projects.  This is not a new point of understanding for us, and thus has been a primary influencer in the investment decisions we make around our product teams.  There are many workflow and feature enhancements around C4R that have not been executed as a result of tradeoffs for that focus, and we continue to work to ensure that Cloud Worksharing services delivered by C4R are the best, most reliable, option relative to the other available options on the market.

 

Specific to the recent outages, if it's any consolation (I'm guessing it's not), it's not been an artifact of load or scalability, but primarily resiliency of Cloud Worksharing services in the face of outages of dependent services.  The past 3 months of investment have been almost exclusively focused around that area, and we hope to deliver that value to subscribers like yourself shortly.  Reglardless of the cause though, what matters is trust and confidence, and like I said back in July 2015, it's up to Autodesk to engender that trust and confidence that goes along with the game-changing collaboration that C4R delivers.

 

2.  Model Corruption

To my knowledge, I know of no situation in today's latest C4R code where model corruption happens as a result of cloud worksharing transactions.  Model corruption, like service availability and reliability, are key to trust and confidence, and we take any such instances extremely seriously across the Revit team.  As such, we'd very much like to investigate the experience you had, as it's contrary to the investment and testing that we do in this area.  Please send me a private message with your support case ID and I will make sure we take a look at what happened.  Additionally, it's a good idea to ensure your teams are running the latest Revit & C4R addon versions for the respective major version you are running on your projects.  I know we had one defect back in February - since fixed - that was resulting in a low incidence or corruption, and perhaps that might be the root cause of your particular issue.

 

We look forward to hearing back from you, and re-building the trust and confidence that's been lost for you recently.

 

Sincerely,

Kyle


Kyle

 

So this post is now about a month old and we are still getting outages at least weekly if not more. Its been down TWICE today alone! So what exactly is Autodesk doing to re-build the trust you have lost over the past 6 months by having so many outages.  Every time users post that the service is out, Autodesk responds with we are working hard to this to make future outages less etc. I even got this response today from another Autodesk employee.  So when will that reliability ACTUALLY come online.

 

On a second note, we have paid lots of money for access to this service, and we expected to see improvements, new features etc. which were promised from the beginning.  From your statement above it looks like all our subscription money is going to support keeping the tool breathing and on life support.  So my investment has not gotten anything new in nearly a year when the A360 tie-in was improved.

 

That said I agree with the OP. The first 18 months - 2 years of service were great. The product is great. The concept is great.  The execution has generally been terrible. 

Message 11 of 14
KyleB_Autodesk
in reply to: chubbard


chubbard wrote:

 

Kyle

 

So this post is now about a month old and we are still getting outages at least weekly if not more. Its been down TWICE today alone! So what exactly is Autodesk doing to re-build the trust you have lost over the past 6 months by having so many outages.  Every time users post that the service is out, Autodesk responds with we are working hard to this to make future outages less etc. I even got this response today from another Autodesk employee.  So when will that reliability ACTUALLY come online.

 

On a second note, we have paid lots of money for access to this service, and we expected to see improvements, new features etc. which were promised from the beginning.  From your statement above it looks like all our subscription money is going to support keeping the tool breathing and on life support.  So my investment has not gotten anything new in nearly a year when the A360 tie-in was improved.

 

That said I agree with the OP. The first 18 months - 2 years of service were great. The product is great. The concept is great.  The execution has generally been terrible. 

 


Chris,

A couple points in response to your questions above:

 

  1. What We Are Doing
    There are a number of work streams in progress right now.  Much of the work is pretty technical, but I'll try my best to describe it in a way that makes sense:

    1. Core Resiliency of C4R Services, and the Foundational Services they Depend On- The reality of cloud products is that behind the curtain things are not always operating normally at some part of the ecosystem of dependent services.  The source of that abnormal operation can be the result of numerous things: bottlenecks at some point in the code, capacity scaling under varying load, intermittent connectivity, among many other things.  Our product teams, across our cloud infrastructure, are working to ensure that our services have the right approaches and architecture to carry out their core operation consistently and with high reliability.  And when they don't, making sure that they self-heal, fail over gracefully to other capacity, or at worst degrade gracefully.  Some of the work here is deliverable in the short-term, while other work is medium-to-long-term.

    2. Soft Dependency of Services on Foundational Services- While teams are working hard to ensure individual services are always operational 24/7/365, degradations or outages at some point in the chain is inevitable.  The trick is to architect the services in a way that such dependencies are "soft" and don't bring down the core products.

      A service like Netflix goes as far as randomly taking out sub-services to ensure their architecture is resilient to such issues, and the end users of the service never see any impact.  Right now C4R has "hard" dependencies on some of our foundational services, as evidenced by many of the outages in the past few months.  We've been working to change that, allowing the "lights to stay on" even if there are issues down below.  The culmination of that investment is slated to deploy later this month, which will be a major step forward.

    3. Service Instrumentation, Health Checks, and Analytics- In support of the work above, we are ensuring at all of our services are constantly logging the results of operations, with breadcrumbs that allow investigation down the road.  With that foundational logging, we can build realtime health checks based on normal operation, which allow us to rapidly understand if something is going wrong.  Kind of like a "Check Engine" light in your car.  In addition to the realtime health checks, we trend the data over time to understand usage patterns, which help us to better schedule capacity to better handle those usage patterns.

      While we have always had some level of these checks across the history of C4R's lifespan, we have built a whole lot more over the past 6 months.  At this point we have notification within minutes of any deviation of behavior of the various operations that make C4R tick.  The same rigor is being applied across the ecosystem of services that either exist today, or are being built for future delivery.

    4. Deployment Pipeline for Services- Since these services are all deployed into what we call the "Production stack" of the services, and being used on a daily basis by our subscribers (C4R serves up ~145k SWC operations a week), we don't do any of the above development work on the Production stack.  Rather, we do that on separate stacks, eventually leading to a deployment to Production.

      In simple terms, the goal is for the non-production stacks to simulate the actual production environment that the new code will eventually inhabit.  This means simulating all possible user operations, across a realistic load, in a reliability environment that mimics what is seen in production.  If this all works properly, our product teams find out about quality issues in the code before that quality issue impacts any subscriber.  We have found gaps in that ability to simulate the production environment, and have been working to close them to ensure any quality issues never hit production.

      In addition, we've been improving the production deployment process to not be noticeable to subscribers, as opposed to requiring a maintenance window & downtime to carry out the deployment.  It's called Blue/Green deployment, and C4R had been doing these deployments for the past 12 months, although technically we switch from "blue" to "green" gradually rather than "throwing the breaker".  That same investment has been underway across the ecosystem, with the goal of ending the hard downtime associated with a maintenance window.  Sometimes, a change is big enough that it requires a maintenance window, which brings me to the final investment to discuss.

    5. Re-Architecting for Cloud Standards- While we've built our services based upon best practices for cloud security, reliability, and operations, we know that the market demands compliance with industry & government standards in order to gain the trust and confidence necessary for our products to be used by markets, companies, or specific project types. SOC 2 is a good example of such a standard.  The work to gain compliance to these standards is not trivial, sometimes requiring architectural changes, and we are working towards compliance for certain standards that we are hearing from the market as critical to adoption.
  2. Reliability vs. Features
    As the Autodesk employee primarily responsible for the prioritization of what the C4R product team focuses on, I 1) understand the many opportunities for workflow and efficiency improvements to C4R and 2)understand the value those improvements would deliver to the market were we to implement them.

    The tradeoff decisions we have made to-date have been based on the basic assertion that a C4R product that delivers the necessary trust and confidence to drive adoption is the most important thing we can do.  No new feature or workflow improvement we can deliver really matters if a team doesn't have that trust and confidence.  Hopefully, the above detail helps to illustrate the result of that assertion.  To be clear though, the sum total of our investments is not simply what's above, there's also some in-development things I can't comment publicly on a forum such as this.

    I'd welcome your opinion, and that of others reading the thread, as to whether we're getting that mix correct.  We strive very hard to get this multi-variable calculus right.

I'm glad to drill into any topic here, within the constraints of what I can publicly comment, and would welcome feedback on the overall conversation at hand.  And finally, I totally get the frustration you are communicating here, and hopefully the above information provides some context as to what we've been working on, and how that leads towards the best outcome.

 

-Kyle



Kyle Bernhardt
Director
Building Design Strategy
Autodesk, Inc.

Message 12 of 14
chubbard
in reply to: KyleB_Autodesk

Kyle,

 

First, thank you  for your frank and honest response.  I understand and suspected many of your points are the issues that are being faced.  However I would have expected that Autodeks would have dealt with these types of issues during the development period, and or, in the early months of the service. I am curious how it ran so well for 12-18 months but then became so unstable. What was the tipping point there? Was that point foreseeable by Autodesk?  

 

I and others really feel this fix is too little too late, given we have had to endure 6 months of regular outages and pay for that privilege.  I see no offer of recompense or other incentives to stay with the service.  6 months is roughly a quarter of the entire service life span of the product. Tellingly not the beginning quarter of its lifespan. 

 

You do make a compelling statement that ADESK has figured out the issues and will be implementing the fix this month and from then on everything should be good.  However that was what we bought into in 2015 only to have it fail at the latter half 2016. So if I buy into your argument pay ADESK another $1200/ user then have it fail 2 years from now for different or similar reasons where does that get me? I know you cannot predict the future, but given that we got here with this system I am more than a little concerned we will get there again and have this conversation all over again.

 

In response to point 2.  Yes it has to be reliable, near 100%, and then new features can be added.  I would argue the issues you raised in point one should have been dealt with in 2015 or earlier and that we should not be here at all.  Your cloud service is EXPENSIVE relative to what others are doing with similar (not Revit you have a monopoly so you can charge what ever you want) but other asset management sites that have some level of complexity generally cost less.  Some are currently working better. I would expect that 1/4-1/3 of my fee is going to innovations not sustainment and that Autodesk should have a parallel roadmap for those innovations that is still progressing even if only on testing servers.  While the service was running we should have started seeing those.  I actually saw very little. 

 

So while I thank you for being honest, non-biased, and factual in your response,  I still have not changed my mind on using the service for the near future. I do truly hope that Autodesk figures out how to manage and properly sell its cloud services but till they do I will be working on other solutions.

Message 13 of 14
KyleB_Autodesk
in reply to: chubbard

Chris,

I few more responses:

 

However I would have expected that Autodeks would have dealt with these types of issues during the development period, and or, in the early months of the service. I am curious how it ran so well for 12-18 months but then became so unstable. What was the tipping point there? Was that point foreseeable by Autodesk? 

 

Good question, but not one with a simple answer.  For each outage we have, there's an intense process of investigation and remediation that's carried out.  There is no one source of root cause, I can say that with confidence.  In some cases, there are lessons learned that could have avoided the issue, and that learning is something that we we either apply to our operational process, or in other cases a development effort that's implemented.

 

You do make a compelling statement that ADESK has figured out the issues and will be implementing the fix this month and from then on everything should be good.  However that was what we bought into in 2015 only to have it fail at the latter half 2016. So if I buy into your argument pay ADESK another $1200/ user then have it fail 2 years from now for different or similar reasons where does that get me? I know you cannot predict the future, but given that we got here with this system I am more than a little concerned we will get there again and have this conversation all over again.

 

I get the frustration by you and your fellow subscribers on this front.  The investments I enumerated previously on this thread, which are the subject of intense focus across our cloud service teams, are targeted to make the issues across the past 6 months not really possible anymore.  Like you say, I can't predict the future, but myself and my colleagues certainly are taking the lessons learned, and our overall increased expertise, forward to handle what the future holds.  I've said many times before on this forum, that the C4R product will never be the success the company wants without the trust and confidence of companies to subscribe, and project teams to adopt.  We understand that, and this is primary driver of our investments.

 

In response to point 2.  Yes it has to be reliable, near 100%, and then new features can be added.  I would argue the issues you raised in point one should have been dealt with in 2015 or earlier and that we should not be here at all.  Your cloud service is EXPENSIVE relative to what others are doing with similar (not Revit you have a monopoly so you can charge what ever you want) but other asset management sites that have some level of complexity generally cost less.  Some are currently working better. I would expect that 1/4-1/3 of my fee is going to innovations not sustainment and that Autodesk should have a parallel roadmap for those innovations that is still progressing even if only on testing servers.  While the service was running we should have started seeing those.  I actually saw very little.

 

I alluded earlier that our sum total of investments are not solely focused on service reliability.  There are investments on workflow improvement and efficiency that are in parallel, they are just not things I can detail on such a public forum.  I can, however, detail some of the consistent feedback we've heard that is driving them:

 

  • You've told us that a high trust, live-linking workflow is great on some projects, with the right team and contract terms, while on other projects it's an adoption blocker.  You've asked for increased control over 1) who can work on specific Models and 2) what versions of a discipline team's Models are available for linking by other discipline teams.

  • The ability to view and comment on Models in the browser is great in all, but it's not capable and efficient enough in its current form for some project teams to displace existing solutions for design review or Sheet Set exchange.  You've asked for better tools for markup, task tracking, and issue resolution.  You've also asked for better ways to aggregate Sheets across models, and sharing those contract documents between project stakeholders.  Finally in this area, you've told us that you'd like a more efficient way to get that data from Revit into the cloud Common Data Environment, where today that only happens when a team member publishes a model version from the Revit UI.
  • You've told us that too many workflows require somebody to fire up an authoring tool like Revit, in order to carry them out.  As capable as Revit is, it;s also a complex tool, and you've told us that this sometimes limits who on the project team can carry out certain tasks, and the efficiency for how the task is completed.

  • We have a lot of great solutions for various high value workflows that a BIM authoring team would need.  Things like visualization, analysis, model clash & coordination, and quantification.  You've told us that existing workflows to leverage those capabilities aren't efficient enough for project data that's "born in the cloud" in a product like C4R.  In addition, the actual team experience isn't always consistent across those existing solutions.

So while I thank you for being honest, non-biased, and factual in your response,  I still have not changed my mind on using the service for the near future. I do truly hope that Autodesk figures out how to manage and properly sell its cloud services but till they do I will be working on other solutions.

 

My goal is not to change your mind.  Only you can do that   :).  Rather, my goal is to be as transparent and direct as possible about what our goals are, and the investments we're making to achieve those goals.  Glad to continue it however long you or others are interested.

 

-Kyle



Kyle Bernhardt
Director
Building Design Strategy
Autodesk, Inc.

Message 14 of 14

Just an update for those who many be following this thread.  We've recently deployed an update to the service that implemented the "soft dependency" architecture that I discussed previously.  This is a major step forward in C4R resiliency efforts, so I wanted to share.

 

-Kyle



Kyle Bernhardt
Director
Building Design Strategy
Autodesk, Inc.

Can't find what you're looking for? Ask the community or share your knowledge.

Post to forums  

Autodesk DevCon in Munich May 28-29th


Autodesk Design & Make Report