I have gone through Reading from Cassandra using Spark Streaming and through tutorial-1 and tutorial-2 links.
Is it fair to say that Cassandra-Spark integration currently does not provide anything out of the box to continuously get the updates from Cassandra and stream them to other systems like HDFS?
By continuously, I mean getting only those rows in a table which have changed (inserted or updated) since the last fetch by Spark. If there are too many such rows, there should be an option to limit the number of rows and the subsequent spark fetch should begin from where it left off. At-least once guarantee is ok but exactly-once would be a huge welcome.
If its not supported, one way to support it could be to have an auxiliary column updated_time in each cassandra-table that needs to be queried by storm and then use that column for queries. Or an auxiliary table per table that contains ID, timestamp of the rows being changed. Has anyone tried this before?
I don't think Apache Cassandra has this functionality out of the box. Internally [for some period of time] it stores all operations on data in sequential manner, but it's per node and it gets compacted eventually (to save space). Frankly, Cassandra's (as most other DB's) promise is to provide latest view of data (which by itself can be quite tricky in distributed environment), but not full history of how data was changing.
So if you still want to have such info in Cassandra (and process it in Spark), you'll have to do some additional work yourself: design dedicated table(s) (or add synthetic columns), take care of partitioning, save offset to keep track of progress, etc.
Cassandra is ok for time series data, but in your case I would consider just using streaming solution (like Kafka) instead of inventing it.
I agree with what Ralkie stated but wanted to propose one more solution if you're tied to C* with this use case. This solution assumes you have full control over the schema and ingest as well. This is not a streaming solution though it could awkwardly be shoehorned into one.
Have you considered using composite key composed of the timebucket along with a murmur_hash_of_one_or_more_clustering_columns % some_int_designed_limit_row_width? In this way, you could set your timebuckets to 1 minute, 5 minutes, 1 hour, etc depending on how "real-time" you need to analyze/archive your data. The murmur hash based off of one or more of the clustering columns is needed to help located data in the C* cluster (and is a terrible solution if you're often looking up specific clustering columns).
For example, take an IoT use case where sensors report in every minute and have some sensor reading that can be represented as an integer.
create table if not exists iottable {
timebucket bigint,
sensorbucket int,
sensorid varchar,
sensorvalue int,
primary key ((timebucket, sensorbucket), sensorid)
} with caching = 'none'
and compaction = { 'class': 'com.jeffjirsa.cassandra.db.compaction.TimeWindowedCompaction' };
Note the use of TimeWindowedCompaction. I'm not sure what version of C* you're using; but with the 2.x series, I'd stay away from DateTieredCompaction. I cannot speak to how well it performs in 3.x. Any any rate, you should test and benchmark extensively before settling on your schema and compaction strategy.
Also note that this schema could result in hotspotting as it is vulnerable to sensors that report more often than others. Again, not knowing the use case it's hard to provide a perfect solution -- it's just an example. If you don't care about ever reading C* for a specific sensor (or column), you don't have to use a clustering column at all and you can simply use a timeUUID or something random for the murmur hash bucketing.
Regardless of how you decide to partition the data, a schema like this would then allow you to use repartitionByCassandraReplica and joinWithCassandraTable to extract the data written during a given timebucket.
Related
I am using ScyllaDb, but I think this also applies to Cassandra since ScyllaDb is compatible with Cassandra.
I have the following table (I got ~5 of this kind of tables):
create table batch_job_conversation (
conversation_id uuid,
primary key (conversation_id)
);
This is used by a batch job to make sure some fields are kept in sync. In the application, a lot of concurrent writes/reads can happen. Once in a while, I will correct the values with a batch job.
A lot of writes can happen to the same row, so it will overwrite the rows. A batch job currently picks up rows with this query:
select * from batch_job_conversation
Then the batch job will read the data at that point and makes sure things are in sync. I think this query is bad because it stresses all the partitions and the node coordinator because it needs to visit ALL partitions.
My question is if it is better for this kind of tables to have a fixed field? Something like this:
create table batch_job_conversation (
always_zero int,
conversation_id uuid,
primary key ((always_zero), conversation_id)
);
And than the query would be this:
select * from batch_job_conversation where always_zero = 0
For each batch job I can use a different partition key. The amount of rows in these tables will be roughly the same size (a few thousand at most). The tables will overwrite the same row probably a lot of times.
Is it better to have a fixed value? Is there another way to handle this? I don't have a logical partition key I can use.
second model would create a LARGE partition and you don't want that, trust me ;-)
(you would do a partition scan on top of large partition, which is worse than original full scan)
(and another advice - keep your partitions small and have a lot of them, then all your cpus will be used rather equally)
first approach is OK - and is called FULL SCAN, BUT
you need to manage it properly
there are several ways, we blogged about it in https://www.scylladb.com/2017/02/13/efficient-full-table-scans-with-scylla-1-6/
and basically it boils down to divide and conquer
also note spark implements full scans too
hth
L
I'm very new to the ETL world and I wish to implement Incremental Data Loading with Cassandra 3.7 and Spark. I'm aware that later versions of Cassandra do support CDC, but I can only use Cassandra 3.7. Is there a method through which I can track the changed records only and use spark to load them, thereby performing incremental data loading?
If it can't be done on the cassandra end, any other suggestions are also welcome on the Spark side :)
It's quite a broad topic, and efficient solution will depend on the amount of data in your tables, table structure, how data is inserted/updated, etc. Also, specific solution may depend on the version of Spark available. One downside of Spark-only method is you can't easily detect deletes of the data, without having a complete copy of previous state, so you can generate a diff between 2 states.
In all cases you'll need to perform full table scan to find changed entries, but if your table is organized specifically for this task, you can avoid reading of all data. For example, if you have a table with following structure:
create table test.tbl (
pk int,
ts timestamp,
v1 ...,
v2 ...,
primary key(pk, ts));
then if you do following query:
import org.apache.spark.sql.cassandra._
val data = spark.read.cassandraFormat("tbl", "test").load()
val filtered = data.filter("""ts >= cast('2019-03-10T14:41:34.373+0000' as timestamp)
AND ts <= cast('2019-03-10T19:01:56.316+0000' as timestamp)""")
then Spark Cassandra Connector will push this query down to the Cassandra, and will read only data where ts is in the given time range - you can check this by executing filtered.explain and checking that both time filters are marked with * symbol.
Another way to detect changes is to retrieve the write time from Cassandra, and filter out the changes based on that information. Fetching of writetime is supported in RDD API for all recent versions of SCC, and is supported in the Dataframe API since release of SCC 2.5.0 (requires at least Spark 2.4, although may work with 2.3 as well). After fetching this information, you can apply filters on the data & extract changes. But you need to keep in mind several things:
there is no way to detect deletes using this method
write time information exists only for regular & static columns, but not for columns of primary key
each column may have its own write time value, in case if there was a partial update of the row after insertion
in most versions of Cassandra, call of writetime function will generate error when it's done for collection column (list/map/set), and will/may return null for column with user-defined type
P.S. Even if you had CDC enabled, it's not a trivial task to use it correctly:
you need to de-duplicate changes - you have RF copies of the changes
some changes could be lost, for example, when node was down, and then propagated later, via hints or repairs
TTL isn't easy to handle
...
For CDC you may look for presentations from 2019th DataStax Accelerate conference - there were several talks on that topic.
I have a single structured row as input with write rate of 10K per seconds. Each row has 20 columns. Some queries should be answered on these inputs. Because most of the queries needs different WHERE, GROUP BY or ORDER BY, The final data model ended up like this:
primary key for table of query1 : ((column1,column2),column3,column4)
primary key for table of query2 : ((column3,column4),column2,column1)
and so on
I am aware of the limit in number of tables in Cassandra data model (200 is warning and 500 would fail)
Because for every input row I should do an insert in every table, the final write per seconds became big * big data!:
writes per seconds = 10K (input)
* number of tables (queries)
* replication factor
The main question: am I on the right path? Is it normal to have a table for every query even when the input rate is already so high?
Shouldn't I use something like spark or hadoop instead of relying on bare datamodel? Or event Hbase instead of Cassandra?
It could be that Elassandra would resolve your problem.
The query system is quite different from CQL, but the duplication for indexing would automatically be managed by Elassandra on the backend. All the columns of one table will be indexed so the Elasticsearch part of Elassandra can be used with the REST API to query anything you'd like.
In one of my tests, I pushed a huge amount of data to an Elassandra database (8Gb) going non-stop and I never timed out. Also the search engine remained ready pretty much the whole time. More or less what you are talking about. The docs says that it takes 5 to 10 seconds for newly added data to become available in the Elassandra indexes. I guess it will somewhat depend on your installation, but I think that's more than enough speed for most applications.
The use of Elassandra may sound a bit hairy at first, but once in place, it's incredible how fast you can find results. It includes incredible (powerful) WHERE for sure. The GROUP BY is a bit difficult to put in place. The ORDER BY is simple enough, however, when (re-)ordering you lose on speed... Something to keep in mind. On my tests, though, even the ORDER BY equivalents was very fast.
I am using cassandra 1.2.15 with ByteOrderedPartitioner in a cluster environment of 4 nodes with 2 replicas. I want to know what are the drawbacks of using the above partitioner in cluster environment? After a long search I found one drawback. I need to know what are the consequences of such drawback?
1) Data will not distribute evenly.
What type of problem will occur if data are not distributed evenly?
Is there is any other drawback with the above partitioner in cluster environment if so, what are the consequences of such drawbacks? Please explain me clearly.
One more question is, Suppose If I go with Murmur3Partitioner the data will distribute evenly. But the order will not be preserved, however this drawback can be overcome with cluster ordering (Second key in the primary keys). Whether my understanding is correct?
As you are using Cassandra 1.2.15, I have found a doc pertaining to Cassandra 1.2 which illustrates the points behind why using the ByteOrderedPartitioner (BOP) is a bad idea:
http://www.datastax.com/documentation/cassandra/1.2/cassandra/architecture/architecturePartitionerBOP_c.html
Difficult load balancing More administrative overhead is required to load balance the cluster. An ordered partitioner
requires administrators to manually calculate partition ranges
(formerly token ranges) based on their estimates of the row key
distribution. In practice, this requires actively moving node
tokens around to accommodate the actual distribution of data once
it is loaded.
Sequential writes can cause hot spots If your application tends to write or update a sequential block of rows at a time, then the
writes are not be distributed across the cluster; they all go to
one node. This is frequently a problem for applications dealing
with timestamped data.
Uneven load balancing for multiple tables If your application has multiple tables, chances are that those tables have different row keys and different distributions of data. An ordered
partitioner that is balanced for one table may cause hot spots and uneven distribution for another table in the same cluster.
For these reasons, the BOP has been identified as a Cassandra anti-pattern. Matt Dennis has a slideshare presentation on Cassandra Anti-Patterns, and his slide about the BOP looks like this:
So seriously, do not use the BOP.
"however this drawback can be overcome with cluster ordering (Second key in the primary keys). Whether my understanding is correct?"
Somewhat, yes. In Cassandra you can dictate the order of your rows (within a partition key) by using a clustering key. If you wanted to keep track of (for example) station-based weather data, your table definition might look something like this:
CREATE TABLE stationreads (
stationid uuid,
readingdatetime timestamp,
temperature double,
windspeed double,
PRIMARY KEY ((stationid),readingdatetime));
With this structure, you could query all of the readings for a particular weather station, and order them by readingdatetime. However, if you queried all of the data (ex: SELECT * FROM stationreads;) the results probably will not be in any discernible order. That's because the total result set will be ordered by the (random) hashed values of the partition key (stationid in this case). So while "yes" you can order your results in Cassandra, you can only do so within the context of a particular partition key.
Also, there have been many improvements in Cassandra since 1.2.15. You should definitely consider using a more recent (2.x) version.
I have the following table (using CQL3):
create table test (
shard text,
tuuid timeuuid,
some_data text,
status text,
primary key (shard, tuuid, some_data, status)
);
I would like to get rows ordered by tuuid. But this is only possible when I restrict shard - I get this is due to performance.
I have shard purely for sharding, and I can potentially restrict its range of values to some small range [0-16) say. Then, I could run a query like this:
select * from test where shard in (0,...,15) order by tuuid limit L;
I may have millions of rows in the table, so I would like to understand the performance characteristics of such a order by query. It would seem like the performance could be pretty bad in general, BUT with a limit clause of some reasonable number (order of 10K), this may not be so bad - i.e. a 16 way merge but with a fairly low limit.
Any tips, advice or pointers into the code on where to look would be appreciated.
Your data is sorted according to your column key. So the performance issue in your merge in your query above does not happen due to the WHERE clause but because of your LIMIT clause, afaik.
Your columns are inserted IN ORDER according to tuuid so there is no performance issue there.
If you are fetching too many rows at once, I recommended creating a test_meta table where you store the latest timeuuid every X-inserts, to get an upper bound on the rows your query will fetch. Then, you can change your query to:
select * from test where shard in (0,...,15) and tuuid > x and tuuid < y;
In short: make use of your column keys and get rid of the limit. Alternatively, in Cassandra 2.0, there will be pagination which will help here, too.
Another issue I stumbled over, you say that
I may have millions of rows in the table
But according to your data model, you will have exactly shard number of rows. This is your row key and - together with the partitioner - will determine the distribution/sharding of your data.
hope that helps!
UPDATE
From my personal experience, cassandra performances quite well during heavy reads as well as writes. If the result sets became too large, I rather experienced memory issues on the receiving/client side rather then timeouts on the server side. Still, to prevent either, I recommend having a look a the upcoming (2.0) pagination feature.
In the meanwhile:
Try to investigate using the trace functionality in 1.2.
If you are mostly reading the "latest" data, try adding a reversed type.
For general optimizations like caches etc, first, read how cassandra handles reads on a node and then, see this tuning guide.