Impact 2024: The Industrial Data and AI Conference for and by Users | Nominate Speakers Now for a Ch...
New→Gathering Interest
Hi, @Kristian Nymoen .Thank you for reporting this. I will create an issue for this, and investigate immediately. As this is a bug you can also create a support ticket for this so you can track the issue through support. FYI @Magdalena Rut Best regards, Knut Vidvei
Hi, Richard!Thank you for an excellent idea. We will take this into our discovery and reach out when we have something concrete. FYI @Magdalena Rut Best regards, Knut
That’s another excellent suggestion, and might deserve it’s own “Product Idea” here on hub. You can add is as a new product idea, or I can create one for you if you’d like. Best regards, Knut
Thank you for the email.This will be valuable input to this work! Best regards, Knut
Hi, Richard.Firs of all, thank you for the feedback and the helpful screenshots. You also have a lot of great suggestions.I can see from your chart, with both a high number of time series and a smaller resolution screen, we have some improvements to how tightly the axes are spaced, and how they behave for you.For you, and other users who might read this post, I’ll highlight a few product features that is helpful in these situations:As you show, it can be a challenge to keep the y-axes visible once you get to a high number of visible time series in a single chart. The first thing is of course as you show, to hide individual time series, unless you need to have 10+ visible to support your analysis.And, as you most probably know, if you need to have a lot visible, you can still hide all the y-axes. This gives the full horizontal real estate to the time series. As for your idea of showing the time series name when hovering over a y-axis is fantastic, and we will add it to “Gathering intere
Hey again, Charts user.Here is a screenshot of what to expect when the re-direct goes into effect shortly.From now you should login to fusion.cognite.com to access Charts.If you try to go to the stand-alone url charts.cogniteapp.com, you will receive this message asking you to go to fusion.cognite.com. Best regards, Knut
Hi,You now have the option to view additional data for the events in Charts. By clicking on the “View details” button you will see when the workorder was created. Since different users have different needs for data, also depending on the situation and type of event, you can also now see all the event data, including the metadata in a searchable way.Does this resolve the need you have for additional information?See below a screenshot example with the “Created at” time highlighted:
Hi, Thank you for raising this. I understand your need to have a better way to filter down the subsets of data that are relevant to people in your company, based on the site they are on or type of data they are interested in.We have seen customers navigate the complexities of multiple sites in different ways. Narrowing down data has typically been thougha) Hierarchies b) Filters and 3) Search.And as we’ve seen, customers want to move from selecting from a long list of data sets -> finding the right data product, starting in a data catalogue or through the data explorer. This concept is maturing in Cognite Data Fusion with data products and Data Catalogue.We look forward to any sessions to solve this in a good way together. Best regards, Knut.
Hi, Ibrahim.We are working our way through the review process, and will have an update for you shortly.Best regards, Knut
Hello!This does not look right. Let me try to re-create what is happening and get back to you.If you want, you can also contact support and we might get access to your chart to troubleshoot directly. All the best, Knut Vidvei
Hi, Udit.I hope we get to resolve your access to the public data.Asset Data Insight is unfortunately no longer part of the public offering. Access to Cognite and the industrial public data is through Cognite Data Fusion, including the data explorer and charts. In addition to accessing it via APIs from 3rd party tools such as Grafana, PowerBI, Postman or through our SDKs. I hope this helps, and apologize for any confusion.Best regards, Knut Vidvei
Hi, it definitely makes sense. Would an additional number next to the time result in percent be sufficient to solve your reporting needs?
Hi, Simen.There is a “Threshold” functionality where you can see e.g. how much time a time series is above, below or between a value for the time frame you are evaluating.Access it by pressing the Threshold icon highlighted in red circle on the right side. The number of time the value is exceeded, and the summary of the time exceeded is updated in the right side panel (square red box in the image). This example shows a week of data, and a threshold under 1.4. Best regards, Knut
Hi, again Ramarao.What you describe is already available and released feature where each user can set which columns are visible, and the order of importance. This is as you say because different people have different backgrounds and needs.The icon is at the top right within each data type. Click the icon in the red box in the image. You can even show metadata as columns here. And you can drag to re-arrange the order of your preferred columns. Today these are stored in the browser, but later we will store them in the user profile of the user so it carries wherever they login.The other request made earlier in this thread is the ability for a company to configure the starting-set of columns per customer project, applied to all users. That set of features are not solved, as we are still investigating the best way to handle such a need.Knut
Hi, We are still looking into how we can solve this best. In the mean time the Domain Experts can set their preferred columns themselves in the UI. For understanding your needs better, could you share the preferred columns domain experts in your company would like to see as default, and their order? Best regards, Knut
Hi, Marius. We are still estimating the efforts, and there is a delay in the approval process with Microsoft, but we aim for 1-2 releases after GA. We will post in this thread when we have better estimates. What is your key use case for using PowerBI with FDM? Best regards, Knut
Hi, Ardash, Thank you for the questions on Cognite Functions. This functionality allows you to deploy and run Python code on Cognite Data Fusion on a schedule or by calling the function.The orchestration is up to the user or system integrator to orchestrate themselves based on your use case, but is not an out-of-the-box functionality in Cognite Data Fusion, as Jason says.What type of orchestration would you need to get more value here? Full documentation is located here, with https://docs.cognite.com/cdf/functions/To call a function you use: call = func.call(data={"assetNo": 0})And to get the responsecall.get_response() Best regards,Knut Vidvei
Great,As for the high time delta, we would have to investigate what's going on here and why that step is jumping 9 hours ahead on that time step. Perhaps we can reach out and try to figure out what's happening here? Knut Vidvei
Hello, Patrick.Thank you for the questions. When you download a time series, and select the granularity (not raw data), the aggregates are calculated according to the description here: https://docs.cognite.com/dev/concepts/aggregation/ This is great feedback, and we understand that you would like more metadata in the csv file. This is definitely something we want to explore further so you get the context you need.Until more of this is available in the csv file, the metadata can be accessed in the side panel by pressing the information (i) at the most right column in the table and choose “metadata” in the segment selector on the top of the side panel.
Hi, Richard.As I understand it, you want to first stack the time series, then lock them so they are not moving the y-axis when you are zooming and panning?The expected behaviour is that after enabling stacking, the y-axes should not re-scale. However there are situations where it will trigger the re-stacking and try to optimize the axes for best view of the data.I’ll investigate with the designers and look into a better pattern to lock the axes when zooming and panning, so that it is easier to follow the data at different zoom levels. Knut
Hi, Rishabh. Thank you for this question. We are working on better tooling around unit types and unit of measure, including conversions. Can you please elaborate a bit on how you would work with this? Would you like to set the unit of measure in the extractor? And is it the same unit as the source unit, or do you want to convert the unit as part of the connector loading into Cognite Data Fusion?And could you give a bit more context with regards to which connector you are using? Best regards,Knut Vidvei
Hi, Benjamin. This is a great suggestion, and I understand the need to not having to move the cursor back and forth but set markers along the time series. We'll keep you posted on the development of this and might ask for your feedback on solution proposals. Knut Vidvei
Hi, Richard. Great question! We are looking into how to make the threshold flow efficient, also considering the coming monitoring and alerting on thresholds.The idea to add multiple time series to a threshold is a great one. Please post if you have other ideas on this. All the best, Knut Vidvei
Already have an account? Login
Enter your username or e-mail address. We'll send you an e-mail with instructions to reset your password.
Sorry, we're still checking this file's contents to make sure it's safe to download. Please try again in a few minutes.
Sorry, our virus scanner detected that this file isn't safe to download.