Update modules/04-data-in-the-cloud/index.ipynb

Co-authored-by: Aimee Barciauskas <[email protected]>

Author: maxrjones

modules/04-data-in-the-cloud/index.ipynb

@@ -73,7 +73,7 @@
     "- Scalability - cloud object storage enables nearly limitless simultaneous access across users/connections, without needing to order or decommission servers or hard-drives\n",
     "- Accessibility - anyone in the world, with proper authorization, can rapidly access data shared on the cloud\n",
     "\n",
-    "There are a couple implications that you should be aware of when working with data on the cloud:\n",
+    "Gotchas: There are a couple of considerations to be aware of when working with data on the cloud:\n",
     "\n",
     "- Pay-as-you-go - Most cloud providers use pay-as-you-go pricing, where you only pay for the storage and services that you use. This can potentially reduce costs, especially upfront costs (e.g., you never need to buy a hard drive). However, **it can be easy to forget about data in storage and continue to pay for it indefinitely**.\n",
     "- Time and cost of bringing data to your computer - Hosting the data on the cloud naturally means it's no longer already near your computer's processing resources. Transporting data from the cloud to your computer is expensive, since most cloud providers charge for any data leaving their network, and slow, since the data needs to travel large distances. The primary solution for this is \"data-proximate computing\" which involves running your code on computing resources in the same cloud location as your data. For example, I commonly use NASA data products that are hosted on AWS servers in the 'us-west-2' region, which corresponds to Oregon in the figure above. Following the \"data-proximate computing\" paradigm, I use AWS compute resources that are also in Oregon when working with those data, rather than downloading data to use the computing resources on my laptop in North Carolina. In addition to \"data-proximate computing\", there are many other ways to make working with data on the cloud cheaper and easier. Let's take a look!"