In a Cloud Monitoring dashboard, you have different timespans you can select. Each of them aggregates the data per minute/hour/day/etc.
An example:
The timestamps that are assigned to the x-axis seem to be in the user's timezone. However, are the aggregates calculated based on UTC, or the user's timezone, or even something different?
The consequence of a user's timezone would be that everyone across the globe would see a different graph. Furthermore, I cannot find any settings related to this.
After looking through the documentation, I cannot find the answer to this. Hence, any insight (preferably with a documentation reference) would help. Thank you!
Related
I am trying to get a historical list of the S&P500 underlying stocks mix. all tickers the dates were added to the S&P500 index mix and the dates tickers were removed from the list. and throughout the years for each period what is the mix. I did some search, doesn't seems to have any luck.
if anyone can provide some good search keywords, or suggest a place to look for would be appreciated
this is something very specific.
I currently use backtrader to work on some data. if there is a systematic way to get the data, please let me know as well.
many thanks.
You can access this data systematically in QuantRocket, via data provider Sharadar:
https://www.quantrocket.com/data/?filter=sharadar
We use Metabase to monitor our KPIs, such as total current active items.
It's a great tool, but when we needed to look at timely trends such as total current active items change by day.
Metabase can't help much, since it only displays what current database has.
I know there are many datawarehousing stacks I can use to store time-series data.
But just curious if there's any framework that can provide a framework to store time-series data of configured KPIs?
What I imagined is like an advanced metabse, the user scenario would be like this:
I can setup metrics I'd like to trace (like DAU, Total Active Items)
The framework automatically saves these metrics by specified time-slot (day, minute, week, month, on user's configuration)
Visualize these data
Would like to have some inputs and thought :D
I'm trying to set up a monitoring dashboard that contains two graphs. One that shows current hour transaction volumes (in 1 minute intervals from current hour start until now) and one that shows current day transaction volumes (in 10 minute intervals from 00:00 until now). I can't seem to find a way to display two different x-axis timelines on the two different panels if I create them on the same dashboard. Is there a way to do what I'm looking for?
I've tried updating the queries themselves, messing with the dashboard settings, and messing with the panel settings but I haven't found what I needed. I'm using Grafana 6.0.0
Just found the answer in the docs: Relative time. With this option you can set a timerange per graph.
A bit too late for this answer, but might find someone who is in need of it though. You could have different time ranges for different panels on the same dashboard. I have a InfluxDB as data source and all I did was to update the Relative Time in the Query options.
In my case, the dashboard has a time range of 30 days but doing this way for a specific panel I was able to show data for the last 24 hours. The value in field Relative time should be positive (-24h will not work here).
Hello ElasticSearchers,
I want to do some practice on sample data set. Can anyone please point me to some freely available data set for practice.
Thanks.
The Global Terrorism Database (GTD) is an open-source database including information on terrorist events around the world from 1970 through 2016 (with annual updates planned for the future). https://www.start.umd.edu/gtd/
The Armed Conflict Location and Event Data Project (ACLED) is a project that collates data on political violence in developing states, from 1997 to the present. https://www.acleddata.com/data/
2 datasets for the same category, covering the same topic and timeline, geo-tagged, a good example for collect, explore and transform, ideal to make a search engine, aggregator, and deduplication exercises.
I am in a bit of dilemma.. in the database if have lots of posts which all have a longitude and latitude attribute.
What I am currently doing is observing the whole post group, and then doing the validation on the device. Which will become very heavy if there are more than 1000 or more posts.
if (location (latitude + longitude)).distance(currentUser.location) < 5000
Is there a way to do the validation directly on the database? And keep in mind that the currentUser.location is relative to the device, so I always would need to construct the location from the post and compare it to the current location of the user.
Any help would be much appreciated.
It's possible if you would use something like geohashes. Then you could filter your locations based on their geohash indexes. Or you can also achieve that by creating your data structure based on geohashes. But there's no need for this since there are libraries which do that.
I've created a library for geo queries, check it out.
Or there's also a good library from firebase itself.