# Too much data?

**URL:** <https://talk.observablehq.com/t/too-much-data/1545>\
**Category:** Help\
**Created:** [January 2, 2019, 4:38pm UTC](https://talk.observablehq.com/t/too-much-data/1545 "2019-01-02T16:38:01Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![katy-li](https://yyz2.discourse-cdn.com/flex030/user_avatar/talk.observablehq.com/katy-li/32/1313_2.png) [@katy-li](https://talk.observablehq.com/u/katy-li)\
**Post date:** [January 2, 2019, 4:38pm UTC](https://talk.observablehq.com/t/too-much-data/1545/1 "2019-01-02T16:38:01Z")

</div>

I have been looking at the posts on downloading or embedding observable notebooks because in the copy of this [Sunburst](https://beta.observablehq.com/@mbostock/d3-zoomable-sunburst) I think I have maxed out space available for my data- and it’s still incomplete.

Since I maxed out the field for my data, I was hoping to download the js and keep working on it, but from what I understand about downloading/embedding the file I still reference a link and don’t download straight code. I was hoping that I could download it and add in remaining data so I could see how the chart may end up looking.

Here is the [link](https://beta.observablehq.com/@katy-li/omni-sunburst) to my forked version. If there is another way I should format my data, or if there is a way to download the code and work with it there?

---

<div class="post-metadata">

**Author:** ![tom](https://yyz2.discourse-cdn.com/flex030/user_avatar/talk.observablehq.com/tom/32/2262_2.png) [@tom](https://talk.observablehq.com/u/tom)\
**Post date:** [January 2, 2019, 5:00pm UTC](https://talk.observablehq.com/t/too-much-data/1545/2 "2019-01-02T17:00:59Z")

</div>

Hi,

Sure thing - you can, and probably should, move the data to a separate file. Check out Mike’s [introduction to data](https://beta.observablehq.com/@mbostock/introduction-to-data) notebook for a guide, and this [fork of your notebook](https://beta.observablehq.com/d/e2475044c297e2da) for an example of how it’s done: I copied the data cell, got rid of `data = (` in the beginning and `)` at the end - so that it’s valid JSON - and then replaced the data cell with a call to `d3.json` to fetch the data from the [GitHub gist I created](https://gist.github.com/tmcw/51b6d5c95d5a55ecfcb7e2b505cd33e9).

Cells can certainly contain a bit of data, but they aren’t intended to be general-purpose storage for long data values, so this one hit the limit. Using GitHub and loading larger data files is the recommended way, and should work really well for this issue.

-Tom