gping VS parquet-format

gping is a souped-up version of the traditional ping utility that graphs network latency for multiple hosts as well as execution time for commands, with the option of custom colors. Our thanks for the suggestion go to fudgecakekistan.

This sounds similar to what I had in our neighborhood. Turned out there was damage on the wire up on the pole, and it started disconnecting at the break point as the wire heated up during the day and would reconnect normal as it cooled in the evening until the next day as the sun came up around 8-10am. I was able to monitor and graph the lost packets and within the terminal with https://github.com/orf/gping during the worst time period. You may go in to test at the wrong time of day and find no damaged wire but try going during the heat up time of day. Ask customers when the worst time is. I’m not a technician but a customer who happens to do IT work. I finally got a good lead tech from the cable provider to check the cable at just the right time of day and they found the break in the cable. There also was a lot of noise in the signal in general at certain frequencies. Our neighborhood now has perfect internet :D

using gping (https://github.com/orf/gping) to wifi router:

Latency Tool: gping

https://github.com/orf/gping - but probably not for 100 IPs..

Right, there's all sorts of metadata and often stats included in any parquet file: https://github.com/apache/parquet-format#file-format

The offsets of said metadata are well-defined (i.e. in the footer) so for S3 / blob storage so long as you can efficiently request a range of bytes you can pull the metadata without having to read all the data.

As i read here https://github.com/apache/parquet-format/blob/master/LogicalTypes.md , they are store in Integer formats and these integers represent the number of days (for Date) or number of milliseconds, microseconds or nanoseconds (for DateTime) since 1970-01-01. This works as expected with the parquet file that written by our ETL tool from internal database --> S3, all Data/DateTime columns are Integers, means that in Glue Job, i have to convert these Integers back to Date/Datetime value to do some transformation on them. But when parquet files are written by Spark, they are Date/DateTime (or TimeStamp to be more concise) format not Integers (i checked by read these files again in other Glue Job) and that make me confused.

Date is confusing with a timezone (UTC or otherwise) and the doco makes no such suggestion.

The Parquet datatypes documentation is pretty clear that there is a flag isAdjustedToUTC to define if the timestamp should be interpreted as having Instant semantics or Local semantics.

https://github.com/apache/parquet-format/blob/master/Logical...

Still no option to include a TZ offset in the data (so the same datum can be interpreted with both Local and Instant semantics) but not bad really.