TERRA ref architecture

[ghost]

Please update this versions of the slides with TERRA-REF consistent design.

https://docs.google.com/presentation/d/1be6d12byKo59On--NZHKcRAoh9K-dOQeI23sUJQcgiA/edit?usp=sharing

I’m not sure what the globus connection is (are you just saying you use globus for transfer?) and how the user interacts with BETYdb (it has a separate GUI, right?)

[jdmaloney]

@rachelshekar Yep, just saying that the transfer protocol is Globus, the detailed path breaks down the bandwidth and service providers that are moving data between the locations. I proof read slide 2 and it looks good to me, I don't see any technical things that need corrected. Awesome diagram too, it looks nice!

[ghost]

@max-zilla : please comment on/edit slides - especially 3 and 4.

[max-zilla]

@rachelshekar I have revised the deck a bit.

• cleaned up slide 1 KSU box
• Revised Slide 3. I duplicated the original just in case, but updated to reflect the Globus pipeline we are going to use for Danforth.

For slide 4, there were some discrepancies between what I interpreted the extractors to be and who was assigned them.

• Jerome working on EnvironmentalLogger -> NetCDF extractor: https://github.com/terraref/computing-pipeline/issues/115
• Xingchen working on Hyperspectral -> NetCDF extractor: https://github.com/terraref/computing-pipeline/issues/81
• Yaping working on metadata geospatial extractor: https://github.com/terraref/computing-pipeline/issues/101
• Pless working on Binary -> JPEG/TIFF conversion extractor: https://github.com/terraref/computing-pipeline/issues/64
• Also the full-field mosaic: https://github.com/terraref/computing-pipeline/issues/85

I didn't have anything yet for the NIR/SWIR portion.

Please let me know what you think of the updates...