Community Discussions

This document outlines a concept that CDF has been under development for the last 2 years. As mentioned during the disclaimer above, please use this Cognite Hub group and share any feedback you have around flexible data modeling.This document is reflecting some very early thinking from the App Dev Journey team, and is a mental model that will likely change overtime with your feedback!What is a Data Model​A data model enables users to customize the shape, structure their expectation of data. It plays a crucial part in building solutions (like data science models, mobile and web apps). But it is also is the core of an ontology, knowledge graphs, or industry standard.There are some crucial reasons why data models are effective for the industrial space. Data modeling enables explicit language, flexible customization, governed iteration, and enhanced accessibility towards data. Let's dive further into each of these qualities of data modeling. Data Model is Explicit​A data model needs to be

Hi everyone! 👋We are thrilled to announce that the next major release of Cognite Data Fusion is just around the corner, launching on June 4th, 2024. This update is packed with exciting new features across our Industrial Tools and Data Operations capabilities.Head over to Product Updates for a sneak peek of what's coming! Tip: hit the subscribe button on Product Ideas, and you'll be notified instantly when our product leaders share product updates.We look forward to your feedback and appreciate your continued contributions!

Hey Data Workflows users!TL;DR If you’re using the workflow execution cancellation endpoint, with the python SDK or without, you will have to update your workflow execution cancellation calls before 29/05/2024. More information below.In our push towards making the Data Workflows API generally available in the June release, we aim to ensure a consistent and expected experience across the API. For this reason, we’ve decided to make a breaking change to the workflow execution cancellation endpoint. Endpoint changesIn summary, the cancellation endpoint now only allows the cancellation of a single execution at a time instead of allowing multiple executions to be cancelled in one call. The updated API documentation can be found here. The python SDK, starting from version 7.42.0, will now point to the new endpoint. An example call to the new endpoint using the Python SDK can be found here.TimelineThe new endpoint is already available for use. To assist with the transition, the old endpoint wi

Hi!We have decided to move the feature from “Integrate” to “Contextualize”. We believe the process of parsing documents is a contextualization process as this is a process that takes places when the data is in CDF. This change should be reflected by the end of the day.Regards,Redza RosliSoftware Engineer in AI in Data Onboarding

We're thrilled to share that CDF user interface is available in 10 new languages (Deutsch 🇩🇪, Español 🇪🇸, Français 🇫🇷, Italiano 🇮🇹, Nederlands🇳🇱, Português🇵🇹, Svenska 🇸🇪, 한국어 🇰🇷, 中文🇨🇳, 日本語 🇯🇵)  to make your experience even more accessible and user-friendly.We've expanded our language support to bring the power of our platform to a global audience. You can now select your preferred language from the upper right corner of your profile by clicking on <Manage account>:Click the avatar on the top rightWhen on the /profile/ page click on the <Language>  left button pick your preferred language from the list:Button reads <Langue> as user has chosen Français language from beforeAfter one selects the language, the page will reload and present the user with the interface in the chosen language.  Your Opinion Matters 🗣️We're committed to ensuring that our multilingual interface exceeds your expectations & we want to hear from you! To do that, we need to

Hey Data Workflows users!We’ve just released highly-requested feature for Data Workflows, namely, being able to retry failed or timed-out executions. This endpoint resumes a previously failed, timed out, or terminated workflow execution by retrying tasks that did not complete successfully. It aims to resume execution activity from the point(s) of failure.Behaviour of the retry operation:Targeted Task Retry: Only retries tasks that have stopped in a terminal state such as CANCELED, FAILED, FAILED_WITH_TERMINAL_ERROR, and TIMED_OUT. Optional tasks are not retried. Subworkflows and Dynamic Tasks: When a failure occurs within a subworkflow or as part of a dynamic task, only the individual nested tasks that failed are retried. The subworkflow or dynamic task container itself is not retried. Retry Limits: Tasks that have reached or exceeded their designated retry limits will not have their retry counts reset to zero. Instead, each retry request permits these tasks a single additional retry.T

When installing pygen in a CDF notebook you may be met with ValueError: Requested 'typing-extensions>=4.10.0; python_version < "3.13"', but typing-extensions==4.7.1 is already installedThis is currently a known bug, which we are working on solving. For now, the workaround is to manually uninstall `typing-extensions` using micropip. The code to do so is documented in the installation of pygen along with other known issues and solutions.

If you ever wondered what each of the different professions in Cognite do, Generative AI has the answer :)  

The recently released version `7.37.0` of the Python-SDK which pygen depends on broke pygen. This is fixed in `0.99.20`. For versions before 0.99.20 you will be met with `ImportError: cannot import name 'ListablePropertyType' ...` when you try to generate an SDK.

@Ragnhild Byrkjeland, @perolssoen, and @Christopher Tannum After using the new scene functionality in CDF, there are several things that I like about the application but also am experiencing some challenges that I believe need to be addressed before we can get end users to leverage the application:the asset layer seems to take a while to load on the backend. Not sure if there’s a way to set this up to load faster? once assets are loaded, the pop-up window that shows the contextualized information is very glitchy - rendering it almost un-useable In general it doesn’t seem like the asset overlay is very reactive to a click of the mouse, but more of a hover over which seems less intuitive for end users It would be great to have a speed slider - like the one that was present in remote - to fine-tune the navigation. This will be particularly helpful when we bring in our subsea models into the scene It would be great if we could set certain attributes to default to be toggled “off” when firs

Hello Community Members,I am delighted to announce that from today, the Cognite Python SDK now has support for Time Series Data Quality Status codes.Using the SDK, you may retrieve data points including their status codes, and filter on status codes if desired. You may also insert data points with status codes.For a full description of all the methods and functionality, please refer to the Python SDK Documentation and our Developer DocumentationWhilst this functionality remains in Beta maturity, we are targeting the upcoming June release for General Availability. Cognite relies on customer feedback as key part of its product maturation process, and customers are warmly encouraged to begin implementation of support for data quality status codes and to provide the valuable feedback necessary to allow us to make this feature Generally Available.Thank you for your continued support! Glen

December 15th 2022 - SLB Offices @ Pune, IndiaOn the week of Dec 12th to Dec 16th, Maureen Byrne (Software Engineer), Maria Fonseca (Technical Support Engineer) and me (Data Engineer) travelled to Pune, India, to deliver the first Cognite Data Fusion Bootcamp for eighteen participants from SLB and Infosys. This training delivery marks a milestone for customer and partners of Cognite: this is the first training where Cognite instructors guide them through an end-to-end deployment of Cognite Data Fusion, including data integrations and solution-building.The bootcamp is an intensive 5-day training program that focuses on a complete deployment, bringing the different steps together with DevOps best practices from CDF bootstrapping to solution deployment, monitoring, and operations. Congratulations to the participants of the first bootcamp for gaining their Cognite Data Fusion Bootcamp Certificate!The three of us were very impressed by how fast the participants adopted DataOps tools and pra

1. Field ‘name’ is required during creating data model in CDF.  But it is not required if create through SDK:data_models = [DataModelApply(space="mySpace",external_id="myDataModel",version="v1")] c.data_modeling.data_models.apply(data_models) 2.  For data models have values in ‘name’ and ‘description’ fields these fields would be cleared up if these parameters are not specified in DataModelApply. These code will clear name and description values (if ‘myDataModel’ had them):data_models = [DataModelApply(space="mySpace",external_id="myDataModel",version="v1")] c.data_modeling.data_models.apply(data_models)These code won’t clear name and description values (if ‘myDataModel’ had them):data_models = [DataModelApply(space="mySpace",external_id="myDataModel",version="v1", name=”myDataModel”, description=”My Description”)]c.data_modeling.data_models.apply(data_models)To fix make changes in cognite-sdk:Set ‘name’ as required for DataModelApply. Fix clearing values from name and description in D

Hi all!The Cognite Data Modelling product team are looking to make an improvement to how properties from views selected in data modelling graph queries are included in the result data set.The documented and desired behaviour is that only view properties that are explicitly selected in the query will be included in the data output. However, there is a known issue with property selection where in some cases a query would return more properties per view than initially selected.The issue manifests such that properties selected on one view will also be included in the result properties of another view in the query. This behaviour only appears in specific cases: the selection should be in the same table expression, the selection should use two or more views that share some container properties, and one or more shared properties (that is, properties mapped to the same container property) should be selected in one of the views but not in the other.As such, the issue does not cause a data leak

We are pleased to announce that cognite-pygen is available in beta. What does Pygen enable? It enables you to write python-code using the concepts in your data model. The code you can write enables you to query CDF and work with the data in your data model. How? Pygen generates a Python-SDK tailored for your data model. Want to know more? https://cognite-pygen.readthedocs-hosted.com/en/latest/what_is_pygen.htmlDisclaimer! In the beta period, we ask you to be careful when using the pygen generated SDK on a live production project.We would love to get feedback on how to improve the usability and robustness of the pygen. Toward General Availability, we will focus on robustifying the generation process.

Hi everyone,We have great news for you. Our team was working hard to make this happen for this release. We have created a smooth integration and transition between Industrial Canvas and Charts.From now on you can bring your charts to your canvases. You can do it in 2 ways: While in Industrial Canvas, click 'Add data' to open the resource selector and select 'Charts' tab as the fifth option from the tabs. Here you can filter and search your charts, select the ones you want in your canvas like any other data type and shazam! your charts appear in your canvas. While navigating through Charts, select your favorite chart and click the 'Actions' button, then locate Industrial Canvas beneath 'Open In' section. Follow the steps by selecting one of your existing canvases or initiate a new one to add your chart to. With this feature you can fully interact with your Charts from the Industrial Canvas:You can update axis ranges Slide left or right to update time window Use shift to select a speci

Hello everyone! I have several binary timeseries that are displayed correctly when I ask for a short period of time: However, if I ask for a longer timespan, like a year, Charts gives me the aggregates, which I believe would be fine on some cases, but in this particular binary case, it does not make any sense. The final user cannot use this information to perform any analysis; any calculations, thresholds, or just “taking a look to see the trend” will not give an accurate result. Is it possible to choose to disable the use of aggregates for cases like this? Thank you in advance.

Hi Team,It would be a good use case to add sequences and SimInt outputs to be added to Canvas since well logs are stored as a sequence in cdf.Also, SimInt outputs are stored as time series outputs. It could be useful to include these as well.Thanks,Akash

 Check out our new comprehensive course on Cognite Data Modeling which is a new addition to the Data Engineer Basics - Transform and Contextualize learning path.In this course, you will learn fundamental concepts, practical applications, and advanced techniques for effectively organizing, querying, and managing industrial data. Whether you're new to data modeling or looking to deepen your understanding, this course will equip you with the knowledge and skills to harness the power of data modeling in Cognite Data Fusion (CDF).Key Concepts:The course covers essential concepts of Data Modeling, such as Data Models, Spaces, Containers, Views, and Instances, as well as its public REST and GraphQL APIs.For a limited time, you can access this course for free. Use the code datamodel-qzb8pin9ukz2w upon registration to the Data Engineer Basics - Transform and Contextualize learning path until 31 March 2024. You can also access five additional courses in the learning path, along with the exam to

Hello everyone.I have a use case that I have been trying to figure and I'm hoping you could help me.We have an equipment, let's say a pump, and we have a timeseries for its flow. We consider that the pump is operating as long as the flow is above a threshold.In PI, if I want to know how long the pump was operating, I can simply use PITimeFilterVal function in Excel to retrieve the amount of time that the timeseries was above the threshold. Me and my team have been trying to do this with Cognite, with no success. We tried transforming the timeseries to a 0 or 1 using the “Threshold” function, and then integrating it, but we have faced some limitations due to the approximations that are intrinsic to the integration function.Unfortunately, I cannot show you the actual data, be we checked a few days where the pump starts the day operating - at 00:00 - and at 2 am our integration gives us a value of 1.84h - when it should be as close as 2h as possible. This difference, even though slight, i

Hi all!Data Workflows (re-named) is now in Public Beta, as part of the latest release of Cognite Data Fusion. See the Product Release Spotlight post for more information.We have also added a new feature for subworkflows which you can read more about here.You can find updated documentation at Cognite Docs and Cognite API spec, replacing all earlier versions. Note that the user interface in Fusion remains in Alpha pending the development of a new version. If you lack access to the user interface, get in touch with a Cognite representative or leave a comment below.Let us know if you have any questions or comments!

Hello group members, “If you were to use CDF as an Industrial Historian, what features relating to versioning, tracking of changes and auditing would you need from our Time Series API”.In an upcoming release, we are planning to introduce support for versioning on Time Series, with the intention to enable our Time Series service to be used in the same way as you would a traditional industrial Historian.I am keen to gather information and insights from you to understand the scenarios where such a capability is required, in order that we build the right thing.  Please feel free to send me a direct message if this is of interest to you!Kind Regards, Glen

If you are one of those, who either have a big vision of maintenance transformation or are working on optimizing one part of the maintenance management process, here is a suggestion on how to optimize to meet ambitious OPEX reduction goals while building sustainable and integrated Maintenance management processes. My name is Aleksandra, and I am a maintenance domain expert in Cognite. Together with my colleagues, we have a mission to empower the maintenance ecosystem with the data and technology so maintenance professionals can efficiently optimize and transform their underlying and fundamental maintenance management processes. Here are some tips on building with the end in mind and avoiding optimization in silos. Optimize using data from multiple systemsMany companies these days are busy with the optimization of their maintenance strategy for equipment and moving it from calendar to condition-based maintenance (CBM), where possible. One way of doing it is to use a centralized conditio

Hello everyone in the group!I’m delighted to announce that in the upcoming release, the Time Series Subscriptions feature will mature to a Generally Available level of maturity.  But…!  The feedback does not stop there.  Please do continue to use this group to ask questions about the feature, request improvements or provide any other form of feedback.In light of the news, I also feel that this group in its current form has served its purpose.  However, we would like the group to continue as we appreciated your engagement on our product developments.  Therefore, we would like to repurpose this group to cover more topics and features relating to the Time Series API, and our other OT services too, such as Events.We hope you continue to be members of the group, and look forward to collaborating with you with our future developments, and thank you so much for your contributions in bringing the Subscriptions feature to GA! 

There’s an issue if you create a label and delete it as you might not be able to reuse the same external_id ever again. Even if you first delete the asset or relationship that the label is conected to first and then delete the label, you end up with a ghost label. With some random exceptions(say in 1 of 15 cases it’s not an issue). It feels like a cache issue, but as we’ve seen over the last half a year or so, the issue seems to persists over several months.For this example I’m using an empty dataset, I’ve created a label, tied it to an asset, deleted the asset and then deleted the label. The external id I’ve used for my label are now locked forever. I then try to create the label, if not found: Resulting in an error This is a bug across the C# and the Python SDK.