I have a data pipeline running every hour, running a HiveCopyActivity to select the past hour's data from DynamoDB into S3. The table I'm selecting from has a hash key VisitorID and range key Timestamp, around 4 million rows and is 7.5GB in size. To reduce the time taken for the job, I created a global secondary index on Timestamp but after monitoring Cloudwatch, it seems that HiveCopyActivity doesn't use the index. I've read through all the relevant AWS documentation but can't find any mention of indexes.
Is there a way to force data pipeline to use an index while filtering like this? If not, are there any alternative applications which could transfer hourly (or any other period) data from DynamoDB to S3?
The DynamoDB EMR Hive adapter currently doesn't support using indexes, unfortunately. You would need to write your own sweeper that scans the index and outputs it to S3 - you can check out https://github.com/awslabs/dynamodb-import-export-tool for some basics to implementing the import/export pipe. The library is essentially a parallel scan framework for sweeping DDB tables.
Related
I am fairly new to AWS and am a bit overwhelmed with the options for a task I have.
Have: I have a dimensional model in an S3 bucket (read only access), that has a folder structure and contains partitioned parquet files. This bucket will be updated daily (+40GB a day), with changes to both dim and fact tables. I need to get this data out of S3, but it's extremely inefficient to set up a boto3 connection and repeatedly pull the entire raw data and continuously check if the data has even been updated.
What I was thinking for a solution: To maintain updated tables in another S3 bucket that I create (likely Athena query outputs), where I can just pull in the updated changes, so that boto can just check if there is data in the new bucket and pull, reducing load.
Considerations:
I need some kind of event notification that triggers the Athena query. I was looking into Lambda or Cloudwatch, but unsure which is better or restraints.
For the fact tables, I need an Athena query that gets the most recent "Last Updated" timestamp from the updated data. And then updates the updated bucket tables to include all the raw data that is greater than the found timestamp.
FYI: I am working with partitioned data, and I am not sure if I can just work with the tables as partitions (part-0000dim-table-3.parquet) or if additional steps are required to work with partitions.
For the dim tables, I need to somehow scan the entire table for changes (dim tables are a combination of SCD 0,1,2)... unsure how best to do this. In the worst case, I could just point the boto3 connection to the raw dim tables whenever the fact tables update.
What AWS APIs, workflows, should I think about using?
I am unclear on the constraints that I could run into with either Lambda, Cloudwatch, Athena step functions, etc. and trying to learn as I go. I am also struggling on how to compare Athena query results across the two buckets.
Thank you very much & if there is any more information that would help, just let me know!!
I am running a simple Athena query as in
SELECT * FROM "logs"
WHERE parse_datetime(requestdatetime,'dd/MMM/yyyy:HH:mm:ss Z')
BETWEEN parse_datetime('2021-12-01:00:00:00','yyyy-MM-dd:HH:mm:ss')
AND
parse_datetime('2021-12-21:19:00:00','yyyy-MM-dd:HH:mm:ss');
However this times out due to the default DML 30 min timeout.
The entries of the path I am querying are a few millions.
Is there a way to address this in Athena or is there a better suited alternative for this purpose?
This is normally solved with partitioning. For data that's organized by date, partition projection is the way to go (versus an explicit partition list that's updated manually or via Glue crawler).
That, of course, assumes that your data is organized by the partition (eg, s3://mybucket/2021/12/21/xxx.csv). If not, then I recommend changing your ingest process as a first step.
You my want to change your ingest process anyway: Athena isn't very good at dealing with a large number of small files. While the tuning guide doesn't give an optimal filesize, I recommend at least a few tens of megabytes. If you're getting a steady stream of small files, use a scheduled Lambda to combine them into a single file. If you're using Firehose to aggregate files, increase the buffer sizes / time limits.
And while you're doing that, consider moving to a columnar format such as Parquet if you're not already using it.
I am looking into different Big Data solutions and have not been able to find a clear answer or documentation on what might be the best approach and frameworks/services to use to address my Big Data use-case.
My Use-case:
I have a data producer that will be sending ~1-2 billion events to a
Kinesis Data Firehose delivery stream daily.
This data needs to be stored in some data lake / data warehouse, aggregated, and then
loaded into DynamoDB for our service to consume the aggregated data
in its business logic.
The DynamoDB table needs to be updated hourly. (hourly is not a hard requirement but we would like DynamoDB to be updated as soon as possible, at the longest intervals of daily updates if required)
The event schema is similar to: customerId, deviceId, countryCode, timestamp
The aggregated schema is similar to: customerId, deviceId, countryCode (the aggregation is on the customerId's/deviceId's MAX(countryCode) for each day over the last 29 days, and then the MAX(countryCode) overall over the last 29 days.
Only the CustomerIds/deviceIds that had their countryCode change from the last aggregation (from an hour ago) should be written to DynamoDB to keep required write capacity units low.
The raw data stored in the data lake / data warehouse needs to be deleted after 30 days.
My proposed solution:
Kinesis Data Firehose delivers the data to a Redshift staging table (by default using S3 as intermediate storage and then using the COPY command to load to Redshift)
An hourly Glue job that:
Drops the 30 day old time-series table and creates a new time-series table for today in Redshift if this is the first job run of a new day
Loads data from staging table to the appropriate time-series table
Creates a view on top of the last 29 days of time-series tables
Aggregates by customerId, deviceId, date, and MAX(CountryCode)
Then aggregates by customerId, deviceId, MAX(countryCode)
Writes the aggregated results to an S3 bucket
Checks the previous hourly Glue job's run aggregated results vs. the current runs aggregated results to find the customerIds/deviceIds that had their countryCode change
Writes the customerIds/deviceIds rows that had their countryCode change to DynamoDB
My questions:
Is Redshift the best storage choice here? I was also considering using S3 as storage and directly querying data from S3 using a Glue job, though I like the idea of a fully-managed data warehouse.
Since our data has a fixed retention period of 30 days, AWS documentation: https://docs.aws.amazon.com/redshift/latest/dg/c_best-practices-time-series-tables.html suggests to use time-series tables and running DROP TABLE on older data that needs to be deleted. Are there other approaches (outside of Redshift) that would make the data lifecycle management easier? Having the staging table, creating and loading into new time-series tables, dropping older time-series tables, updating the view to include the new time-series table and not the one that was dropped could be error prone.
What would be an optimal way to find the the rows (customerId/deviceId combinations) that had their countryCode change since the last aggregation? I was thinking the Glue job could create a table from the previous runs aggregated results S3 file and another table from the current runs aggregated results S3 file, run some variation of a FULL OUTER JOIN to find the rows that have different countryCodes. Is there a better approach here that I'm not aware of?
I am a newbie when it comes to Big Data and Big Data solutions so any and all input is appreciated!
tldr: Use step functions, not Glue. Use Redshift Spectrum with data in S3. Otherwise you overall structure looks on track.
You are on the right track IMHO but there are a few things that could be better. Redshift is great for sifting through tons of data and performing analytics on it. However I'm not sure you want to COPY the data into Redshift if all you are doing is building aggregates to be loaded into DDB. Do you have other analytic workloads being done that will justify storing the data in Redshift? Are there heavy transforms being done between the staging table and the time series event tables? If not you may want to make the time series tables external - read directly from S3 using Redshift Spectrum. This could be a big win as the initial data grouping and aggregating is done in the Spectrum layer in S3. This way the raw data doesn't have to be moved.
Next I would advise not using Glue unless you have a need (transform) that cannot easily be done elsewhere. I find Glue to require some expertise to get to do what you want and it sounds like you would just be using it for a data movement orchestrator. If this impression is correct you will be better off with a step function or even a data pipeline. (I've wasted way too much time trying to get Glue to do simple things. It's a powerful tool but make sure you'll get value from the time you will spend on it.)
If you are only using Redshift to do these aggregations and you go the Spectrum route above you will want to get as small a cluster as you can get away with. Redshift can be pricy and if you don't use its power, not cost effective. In this case you can run the cluster only as needed but Redshift boot up times are not fast and the smallest clusters are not expensive. So this is a possibility but only in the right circumstances. Depending on how difficult the aggregation is that you are doing you might want to look at Athena. If you are just running a few aggregating queries per hour then this could be the most cost effective approach.
Checking against the last hour's aggregations is just a matter of comparing the new aggregates against the old which are in S3. This is easily done with Redshift Spectrum or Athena as they can makes files (or sets of files) the source for a table. Then it is just running the queries.
In my opinion Glue is an ETL tool that can do high power transforms. It can do a lot of things but is not my first (or second) choice. It is touchy, requires a lot of configuration to do more than the basics, and requires expertise that many data groups don't have. If you are a Glue expert, knock you self out; If not, I would avoid.
As for data management, yes you don't want to be deleting tons of rows from the beginning of tables in Redshift. It creates a lot of data reorganization work. So storing your data in "month" tables and using a view is the right way to go in Redshift. Dropping tables doesn't create this housekeeping. That said if you organize you data in S3 in "month" folders then unneeded removing months of data can just be deleting these folders.
As for finding changing country codes this should be easy to do in SQL. Since you are comparing aggregate data to aggregate data this shouldn't be expensive either. Again Redshift Spectrum or Athena are tools that allow you to do this on S3 data.
As for being a big data newbie, not a worry, we all started there. The biggest difference from other areas is how important it is to move the data the fewest number of times. It sounds like you understand this when you say "Is Redshift the best storage choice here?". You seem to be recognizing the importance of where the data resides wrt the compute elements which is on target. If you need the horsepower of Redshift and will be accessing the data over and over again then the Redshift is the best option - The data is moved once to a place where the analytics need to run. However, Redshift is an expensive storage solution - it's not what it is meant to do. Redshift Spectrum is very interesting in that the initial aggregations of data is done in S3 and much reduced partial results are sent to Redshift for completion. S3 is a much cheaper storage solution and if your workload can be pattern-matched to Spectrum's capabilities this can be a clear winner.
I want to be clear that you have only described on area where you need a solution and I'm assuming that you don't have other needs for a Redshift cluster operating on the same data. This would change the optimization point.
I am trying to use AWS Athena to provide analytics for an existing platform. Currently the flow looks like this:
Data is pumped into a Kinesis Firehose as JSON events.
The Firehose converts the data to parquet using a table in AWS Glue and writes to S3 either every 15 mins or when the stream reaches 128 MB (max supported values).
When the data is written to S3 it is partitioned with a path /year=!{timestamp:yyyy}/month=!{timestamp:MM}/day=!{timestamp:dd}/...
An AWS Glue crawler update a table with the latest partition data every 24 hours and makes it available for queries.
The basic flow works. However, there are a couple of problems with this...
The first (and most important) is that this data is part of a multi-tenancy application. There is a property inside each event called account_id. Every query that will ever be issued will be issued by a specific account and I don't want to be scanning all account data for every query. I need to find a scalable way query only the relevant data. I did look into trying to us Kinesis to extract the account_id and use it as a partition. However, this currently isn't supported and with > 10,000 accounts the AWS 20k partition limit quickly becomes a problem.
The second problem is file size! AWS recommend that files not be < 128 MB as this has a detrimental effect on query times as the execution engine might be spending additional time with the overhead of opening Amazon S3 files. Given the nature of the Firehose I can only ever reach a maximum size of 128 MB per file.
With that many accounts you probably don't want to use account_id as partition key for many reasons. I think you're fine limits-wise, the partition limit per table is 1M, but that doesn't mean it's a good idea.
You can decrease the amount of data scanned significantly by partitioning on parts of the account ID, though. If your account IDs are uniformly distributed (like AWS account IDs) you can partition on a prefix. If your account IDs are numeric partitioning on the first digit would decrease the amount of data each query would scan by 90%, and with two digits 99% – while still keeping the number of partitions at very reasonable levels.
Unfortunately I don't know either how to do that with Glue. I've found Glue very unhelpful in general when it comes to doing ETL. Even simple things are hard in my experience. I've had much more success using Athena's CTAS feature combined with some simple S3 operation for adding the data produced by a CTAS operation as a partition in an existing table.
If you figure out a way to extract the account ID you can also experiment with separate tables per account, you can have 100K tables in a database. It wouldn't be very different from partitions in a table, but could be faster depending on how Athena determines which partitions to query.
Don't worry too much about the 128 MB file size rule of thumb. It's absolutely true that having lots of small files is worse than having few large files – but it's also true that scanning through a lot of data to filter out just a tiny portion is very bad for performance, and cost. Athena can deliver results in a second even for queries over hundreds of files that are just a few KB in size. I would worry about making sure Athena was reading the right data first, and about ideal file sizes later.
If you tell me more about the amount of data per account and expected life time of accounts I can give more detailed suggestions on what to aim for.
Update: Given that Firehose doesn't let you change the directory structure of the input data, and that Glue is generally pretty bad, and the additional context you provided in a comment, I would do something like this:
Create an Athena table with columns for all properties in the data, and date as partition key. This is your input table, only ETL queries will be run against this table. Don't worry that the input data has separate directories for year, month, and date, you only need one partition key. It just complicates things to have these as separate partition keys, and having one means that it can be of type DATE, instead of three separate STRING columns that you have to assemble into a date every time you want to do a date calculation.
Create another Athena table with the same columns, but partitioned by account_id_prefix and either date or month. This will be the table you run queries against. account_id_prefix will be one or two characters from your account ID – you'll have to test what works best. You'll also have to decide whether to partition on date or a longer time span. Dates will make ETL easier and cheaper, but longer time spans will produce fewer and larger files, which can make queries more efficient (but possibly more expensive).
Create a Step Functions state machine that does the following (in Lambda functions):
Add new partitions to the input table. If you schedule your state machine to run once per day it can just add the partition that correspond to the current date. Use the Glue CreatePartition API call to create the partition (unfortunately this needs a lot of information to work, you can run a GetTable call to get it, though. Use for example ["2019-04-29"] as Values and "s3://some-bucket/firehose/year=2019/month=04/day=29" as StorageDescriptor.Location. This is the equivalent of running ALTER TABLE some_table ADD PARTITION (date = '2019-04-29) LOCATION 's3://some-bucket/firehose/year=2019/month=04/day=29' – but doing it through Glue is faster than running queries in Athena and more suitable for Lambda.
Start a CTAS query over the input table with a filter on the current date, partitioned by the first character(s) or the account ID and the current date. Use a location for the CTAS output that is below your query table's location. Generate a random name for the table created by the CTAS operation, this table will be dropped in a later step. Use Parquet as the format.
Look at the Poll for Job Status example state machine for inspiration on how to wait for the CTAS operation to complete.
When the CTAS operation has completed list the partitions created in the temporary table created with Glue GetPartitions and create the same partitions in the query table with BatchCreatePartitions.
Finally delete all files that belong to the partitions of the query table you deleted and drop the temporary table created by the CTAS operation.
If you decide on a partitioning on something longer than date you can still use the process above, but you also need to delete partitions in the query table and the corresponding data on S3, because each update will replace existing data (e.g. with partitioning by month, which I would recommend you try, every day you would create new files for the whole month, which means that the old files need to be removed). If you want to update your query table multiple times per day it would be the same.
This looks like a lot, and looks like what Glue Crawlers and Glue ETL does – but in my experience they don't make it this easy.
In your case the data is partitioned using Hive style partitioning, which Glue Crawlers understand, but in many cases you don't get Hive style partitions but just Y/M/D (and I didn't actually know that Firehose could deliver data this way, I thought it only did Y/M/D). A Glue Crawler will also do a lot of extra work every time it runs because it can't know where data has been added, but you know that the only partition that has been added since yesterday is the one for yesterday, so crawling is reduced to a one-step-deal.
Glue ETL is also makes things very hard, and it's an expensive service compared to Lambda and Step Functions. All you want to do is to convert your raw data form JSON to Parquet and re-partition it. As far as I know it's not possible to do that with less code than an Athena CTAS query. Even if you could make the conversion operation with Glue ETL in less code, you'd still have to write a lot of code to replace partitions in your destination table – because that's something that Glue ETL and Spark simply doesn't support.
Athena CTAS wasn't really made to do ETL, and I think the method I've outlined above is much more complex than it should be, but I'm confident that it's less complex than trying to do the same thing (i.e. continuously update and potentially replace partitions in a table based on the data in another table without rebuilding the whole table every time).
What you get with this ETL process is that your ingestion doesn't have to worry about partitioning more than by time, but you still get tables that are optimised for querying.
I have a web app that needs to send reports on its usage, I want to use Amazon RedShift as a data warehouse for that purpose,
How should i collect the data ?
Every time, the user interact with my app, i want to report that.. so when should i write the files to S3 ? and how many ?
What i mean is:
- If do not send the info immediately, then I might lose it as a result of a connection lost, or from some bug in my system while its been collected and get ready to be sent to S3...
- If i do write files to S3 on each user interaction, i will end up with hundreds of files (on each file has minimal data), that need to be managed, sorted, deleted after been copied to RedShift.. that dose not seems like a good solution .
What am i missing? Should i use DynamoDB instead, Should i use simple insert into Redshift instead !?
If i do need to write the data to DynamoDB, should i delete the hold table after been copied .. what are the best practices ?
On any case what are the best practices to avoid data duplication in RedShift ?
Appreciate the help!
It is preferred to aggregate event logs before ingesting them into Amazon Redshift.
The benefits are:
You will use the parallel nature of Redshift better; COPY on a set of larger files in S3 (or from a large DynamoDB table) will be much faster than individual INSERT or COPY of a small file.
You can pre-sort your data (especially if the sorting is based on event time) before loading it into Redshift. This is also improve your load performance and reduce the need for VACUUM of your tables.
You can accumulate your events in several places before aggregating and loading them into Redshift:
Local file to S3 - the most common way is to aggregate your logs on the client/server and every x MB or y minutes upload them to S3. There are many log appenders that are supporting this functionality, and you don't need to make any modifications in the code (for example, FluentD or Log4J). This can be done with container configuration only. The down side is that you risk losing some logs and these local log files can be deleted before the upload.
DynamoDB - as #Swami described, DynamoDB is a very good way to accumulate the events.
Amazon Kinesis - the recently released service is also a good way to stream your events from the various clients and servers to a central location in a fast and reliable way. The events are in order of insertion, which makes it easy to load it later pre-sorted to Redshift. The events are stored in Kinesis for 24 hours, and you can schedule the reading from kinesis and loading to Redshift every hour, for example, for better performance.
Please note that all these services (S3, SQS, DynamoDB and Kinesis) allow you to push the events directly from the end users/devices, without the need to go through a middle web server. This can significantly improve the high availability of your service (how to handle increased load or server failure) and the cost of the system (you only pay for what you use and you don't need to have underutilized servers just for logs).
See for example how you can get temporary security tokens for mobile devices here: http://aws.amazon.com/articles/4611615499399490
Another important set of tools to allow direct interaction with these services are the various SDKs. For example for Java, .NET, JavaScript, iOS and Android.
Regarding the de-duplication requirement; in most of the options above you can do that in the aggregation phase, for example, when you are reading from a Kinesis stream, you can check that you don't have duplications in your events, but analysing a large buffer of events before putting into the data store.
However, you can do this check in Redshift as well. A good practice is to COPY the data into a staging tables and then SELECT INTO a well organized and sorted table.
Another best practice you can implement is to have a daily (or weekly) table partition. Even if you would like to have one big long events table, but the majority of your queries are running on a single day (the last day, for example), you can create a set of tables with similar structure (events_01012014, events_01022014, events_01032014...). Then you can SELECT INTO ... WHERE date = ... to each of this tables. When you want to query the data from multiple days, you can use UNION_ALL.
One option to consider is to create time series tables in DynamoDB where you create a table every day or week in DynamoDB to write every user interaction. At the end of the time period (day, hour or week), you can copy the logs on to Redshift.
For more details, on DynamoDB time series table see this pattern: http://docs.aws.amazon.com/amazondynamodb/latest/developerguide/GuidelinesForTables.html#GuidelinesForTables.TimeSeriesDataAccessPatterns
and this blog:
http://aws.typepad.com/aws/2012/09/optimizing-provisioned-throughput-in-amazon-dynamodb.html
For Redshift DynamoDB copy: http://docs.aws.amazon.com/amazondynamodb/latest/developerguide/RedshiftforDynamoDB.html
Hope this helps.
Though there is already an accepted answer here, AWS launched a new service called Kinesis Firehose which handles the aggregation according to user defined intervals, a temporary upload to s3 and the upload (SAVE) to redshift, retries and error handling, throughput management,etc...
This is probably the easiest and most reliable way to do so.
You can write data to CSV file on local disk and then run Python/boto/psycopg2 script to load data to Amazon Redshift.
In my CSV_Loader_For_Redshift I do just that:
Compress and load data to S3 using boto Python module and multipart upload.
conn = boto.connect_s3(AWS_ACCESS_KEY_ID,AWS_SECRET_ACCESS_KEY)
bucket = conn.get_bucket(bucket_name)
k = Key(bucket)
k.key = s3_key_name
k.set_contents_from_file(file_handle, cb=progress, num_cb=20,
reduced_redundancy=use_rr )
Use psycopg2 COPY command to append data to Redshift table.
sql="""
copy %s from '%s'
CREDENTIALS 'aws_access_key_id=%s;aws_secret_access_key=%s'
DELIMITER '%s'
FORMAT CSV %s
%s
%s
%s;""" % (opt.to_table, fn, AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY,opt.delim,quote,gzip, timeformat, ignoreheader)
Just being a little selfish here and describing exactly what Snowplow ,an event analytics platform does. They use this awesome unique way of collecting event logs from the client and aggregating it on S3.
They use Cloudfront for this. What you can do is, host a pixel in one of the S3 buckets and put that bucket behind a CloudFront distribution as an origin. Enable logs to an S3 bucket for the same CloudFront.
You can send logs as url parameters whenever you call that pixel on your client (similar to google analytics). These logs can then be enriched and added to Redshift database using Copy.
This solves the purpose of aggregation of logs. This setup will handle all of that for you.
You can also look into Piwik which is an open source analytics service and see if you can modify it specific to your needs.