I have an application (built in Meteor) that provides some ad hoc reporting capabilities to the end user. I have built up that functionality by using the aggregation pipeline to produce the results for a given query. This makes it extremely fast and I was using $out to push the results right into a results table.
The results table included a queryID, which the client used to figure out which were the correct results.
Unfortunately, as you may know (and I discovered), that doesn't work so well once you have more than one user running reports at a time because $out deletes the whole results table before pushing the new query in.
I see three possible workarounds:
Run the aggregation, but manually push the results into the results collection
$out the results into a temporary collection (dynamically named to avoid conflicts) and then manually copy the results from there into results collection, immediately dropping the temporary one. This made some sense when I thought I could use copyTo(), but that doesn't appear possible within Meteor, so I think this option doesn't make much sense relative to #1 in this case.
$out the results into a temporary collection (dynamically named to avoid conflicts) and have the client pull its results directly from there. I would then periodically drop the extra collections after say 24 hours (like I do with specific query results in the main collection today).
#3 would be the fastest by far - the time it takes to manually copy rows dwarfs the time it takes the queries to run. But I'm concerned about the impact of creating and dropping so many collections.
We're not talking millions of users here, but if an average of 500 users a day were each running 10-20 reports, there could be an additional 5-10k collections in the database at any one time. That seems like a lot. Perhaps I could be smarter about cleaning them up somehow, though I can't just immediately remove them because a user might want to have multiple tabs open with different reports. Even still, we're potentially talking about hundreds to thousands of collections.
Is that going to be a problem?
Are there other approaches I should consider instead?
Other recommendations?
Thanks!
Dropping a collection in mongoDB is very efficient operation, anyway much more efficient than deleting some documents in a larger collection.
Maximum number of collections is quite high, only limited by namespace namespace in MMAPv1 while no hard limit exists in wiretiger engine.
So I would favor your solution #3.
Some improvements/alternatives you can think:
Consider creating the collections in a separated database (say per day) then you can drop the entire database in a single operation without having to drop individual collections.
Use an endpoint for the result set, cash the results then drop the $out collection. Let cache handle user requirements and only rerun the aggregation if cache has expired or something.
This kind of activity is done very easily in relational databases such as mysql or pgsql. You might consider synchronising your data to a separate relational database for the purposes of reporting.
There is a package https://github.com/perak/mysql-shadow which claims to provide synchronisation. I played with it and it didn't work perfectly, although doing just one way sync is more likely to succeed.
The other option is to use Graphql over a mongo/mysql hybrid database which can be done with the Apollo stack http://www.apollodata.com/
Related
We want to query an entire live production MongoDB collection (v2.6, around 500GB of data on around 70M documents).
We're wondering what's the best approach for this:
A single query with no filtering to open a cursor and get documents in batches of 5/6k
Iterate with pagination, using a logic of find().limit(5000).skip(currentIteration * 5000)
We're unsure what's the best practice and will yield the best results with minimum impact on performance.
I would go with 1. & 2. mixed if possible: Iterate over your huge dataset in pages but access those pages by querying instead of skipping over them as this may be costly as also pointed out by the docs.
The cursor.skip() method is often expensive because it requires the
server to walk from the beginning of the collection or index to get
the offset or skip position before beginning to return results. As the
offset (e.g. pageNumber above) increases, cursor.skip() will become
slower and more CPU intensive. With larger collections, cursor.skip()
may become IO bound.
So if possible build your pages on an indexed field and process those batches of data with an according query range.
The brutal way
Generally speaking, most drivers load batches of documents anyway. So your languages equivalent of
var docs = db.yourcoll.find()
docs.forEach(
function(doc){
//whatever
}
)
will actually just create a cursor initially, and will then, when the current batch is close to exhaustion, load a new batch transparently. So doing this pagination manually while planning to access every document in the collection will have little to no advantage, but hold the overhead of multiple queries.
As for ETL, manually iterating over the documents to modify and then store them in a new instance does under most circumstances not seem reasonable to me, as you basically reinvent the wheel.
Alternate approach
Generally speaking, there is no one-size-fits all "best" way. The best way is the one that best fits your functional and non-functional requirements.
When doing ETL from MongoDB to MongoDB, I usually proceed as follows:
ET…
Unless you have very complicated transformations, MongoDB's aggregation framework is a surprisingly capable ETL tool. I use it regularly for that purpose and have yet to find a problem not solvable with the aggregation framework for in-MongoDB ETL. Given the fact that in general each document is processed one by one, the impact on your production environment should be minimal, if noticeable at all. After you did your transformation, simply use the $out stage to save the results in a new collection.
Even collection spanning transformations can be achieved, using the $lookup stage.
…L
After you did the extract and transform on the old instance, for loading the data to the new MongoDB instance, you have several possibilities:
Create a temporary replica set, consisting of the old instance, the new instance and an arbiter. Make sure your old instance becomes primary, do the ET part, have the primary step down so your new instance becomes primary and remove the old instance and the arbiter from the replica set. The advantage is that you facilitate MongoDB's replication mechanics to get the data from your old instance to your new instance, without the need to worry about partially executed transfers and such. And you can use it the other way around: Transfer the data first, make the new instance the primary, remove the other members from the replica set perform your transformations and remove the "old" data, then.
Use db.CloneCollection(). The advantage here is that you only transfer the collections you need, at the expense of more manual work.
Use db.cloneDatabase() to copy over the entire DB. Unless you have multiple databases on the original instance, this method has little to now advantage over the replica set method.
As written, without knowing your exact use cases, transformations and constraints, it is hard to tell which approach makes the most sense for you.
MongoDB 3.4 support Parallel Collection Scan. I never tried this myself yet. But looks interesting to me.
This will not work on sharded clusters. If we have parallel processing setup this will speed up the scanning for sure.
Please see the documentation here: https://docs.mongodb.com/manual/reference/command/parallelCollectionScan/
I have a highly normalized data model with me. Currently I'm using manual referencing by storing the _id and running sequential queries to fetch details from the deepest collection.
The referencing is one-way and the flow has around 5-6 collections. For one particular use case, I'm having to query down to the deepest collection by querying subsequent "_id" from the higher level collections. So technically I'm hitting the database every time I run a
db.collection_name.find(_id: ****).
My prime goal is to optimize the read without hugely affecting the atomicity of the other collections. I have read about de-normalization and it does not make sense to me because I want to keep an option for changing the cardinality down the line and hence want to maintain a separate collection altogether.
I was initially thinking of using MapReduce to do an aggregation from the back and have a collection primarily for the particular use-case. But well even that does not sound that good.
In a relational db, I would be breaking the query in sub-queries and performing a join to get the data sets that intersect from the initial results. Since mongodb does not support joins, I'm having a tough time figuring anything out.
Please help if you have faced anything like this earlier or have any idea how to resolve it.
Denormalize your data.
MongoDB does not do JOIN's - period.
There is no operation on the database which gets data from more than one collection. Not find(), not aggregate() and not MapReduce. When you need to puzzle your data together from more than one collection, there is no other way than doing it on the application layer. For that reason you should organize your data in a way that any common and performance-relevant query can be resolved by querying just a single collection.
In order to do that you might have to create redundancies and transitive dependencies. This is normal in MongoDB.
When this feels "dirty" to you, then you should either accept the fact that your performance will be sub-optimal or use a different kind of database, like a classic relational database or a graph database.
I'm storing two collections in a MongoDB database:
==Websites==
id
nickname
url
==Checks==
id
website_id
status
I want to display a list of check statuses with the appropriate website nickname.
For example:
[Google, 200] << (basically a join in SQL-world)
I have thousands of checks and only a few websites.
Which is more efficient?
Store the nickname of the website within the "check" directly. This means if the nickname is ever changed, I'll have to perform a mass update of thousands of documents.
Return a multidimensional array where the site ID is the key and the nickname is the value. This is to be used when iterating through the list of checks.
I've read that #1 isn't too bad (in the NoSQL) world and may, in fact, be preferred? True?
If it's only a few websites I'd go with option 1 - not as clean and normalized as in the relational/SQL world but it works and much less painful than trying to emulate joins with MongoDB. The thing to remember with MongoDB or any other NoSQL database is that you are generally making some kind of trade off - nothing is for free. I personally really value the schema-less document oriented data design and for the applications I use it for I readily make the trade-offs (like no joins and transactions).
That said, this is a trade-off - so one thing to always be asking yourself in this situation is why am I using MongoDB or some other NoSQL database? Yes, it's trendy and "hot", but I'd make certain that what you are doing makes sense for a NoSQL approach. If you are spending a lot of time working around the lack of joins and foreign keys, no transactions and other things you're used to in the SQL world I'd think seriously about whether this is the best fit for your problem.
You might consider a 3rd option: Get rid of the Checks collection and embed the checks for each website as an array in each Websites document.
This way you avoid any JOINs and you avoid inconsistencies, because it is impossible for a Check to exist without the Website it belongs to.
This, however, is only recommended when the checks array for each document stays relatively constant over time and doesn't grow constantly. Rapidly growing documents should be avoided in MongoDB, because everytime a document doubles its size, it is moved to a different location in the physical file it is stored in, which slows down write-operations. Also, MongoDB has a 16MB limit per document. This limit exists mostly to discourage growing documents.
You haven't said what a Check actually is in your application. When it is a list of tasks you perform periodically and only make occasional changes to, there would be nothing wrong with embedding. But when you collect the historical results of all checks you ever did, I would rather recommend to put each result(set?) in an own document to avoid document growth.
Is there a way to use the cursor returned by the previous query as a value for $in in the next query? For example, something like this:
var users = db.user.find({state:1})
var offers = db.offer.find({user:{$in:users}})
I think this can reduce the traffic between mongodb and client in case the client doesn't need user information at all, just offers. Am i wrong?
Basically you want to do a join between two collections which Mongo doesn't support. You can reduce the amount of data being transferred from the server by limiting the fields returned from the first query to only the unique user information (i.e. the _id) that you need to get data from the offers collection.
If you really just want to make one query then you should store more information in the offers collection. For example, if you're trying to find offers for active users then you would store the active state of the user in the offers collection.
To work from your comment:
Yes, that's why I used tag 'join' in a question. The idea is that I
can make a first query more сomplex using a bunch of fields and
regexes without storing user data in other collections except
references. In these cases I always have to perform two consecutive
queries, but transfering of the results of the first query is not
necessary neither for me nor for the mongodb itself. I just want to
understand could it be done now, will it be possible to do so in the
future or it cannot be implemented for some technical reasons
As far as I understand it there is no immediate hurry to make this possible. Also the way it is coded atm will make this quite a big change to the way cursors work and are defined. A change big enough to possibly cause implementation breaks for other people. It is really a case of whether to set safe for inserts and updates for all future drivers. It is recognised that safe should be default but this will break implementation for other people who expect it the other way around.
It is rather inefficient if you don't require the results of the first query at all however since most networks are prepped with high traffic in mind and the traffic is cheap there hasn't been a demand to make it able to do chained queries server side in the cursor.
However subselects (which this basically is, it is selecting a set of rows based upon a sub selection of previous rows) have been on mongodb-user a couple of times and there might even be a JIRA for it somewhere, if not might be useful to make one.
As for doing it right now: there is no way.
What is an example of data that is "predistilled or aggregated in runtime"? (And why isn't MongoDB very good with it?)
This is a quote from the MongoDB docs:
Traditional Business Intelligence. Data warehouses are more suited to new, problem-specific BI databases. However note that MongoDB can work very well for several reporting and analytics problems where data is pre-distilled or aggregated in runtime -- but classic, nightly batch load business intelligence, while possible, is not necessarily a sweet spot.
Let's take something simple like counting clicks. There are a few ways to report on clicks.
Store the clicks in a single place. (file, database table, collection) When somebody wants stats, you run a query on that table and aggregate the results. Of course, this doesn't scale very well, so typically you use...
Batch jobs. Store your clicks as in #1, but only summarize them every 5 minutes or so. When people want to query the summary table. Note that "clicks" may have millions of rows, but "summary" may only have a few thousand rows, so it's much quicker to query.
Count the clicks in real-time. Every time there's a click you increment a counter somewhere. Typically this means incrementing the "summary" table(s).
Now most big systems use #2. There are several systems that are very good for this specifically (see Hadoop).
#3 is difficult to do with SQL databases (like MySQL), because there's a lot of disk locking happening. However, MongoDB isn't constantly locking the disk and tends to have much better write throughput.
So MongoDB ends up being very good for such "real-time counters". This is what they mean by predistilled or aggregated in runtime.
But if MongoDB has great write throughput, shouldn't it be good at doing batch jobs?
In theory, this may be true and MongoDB does support Map/Reduce. However, MongoDB's Map/Reduce is currently quite slow and not on par with other Map/Reduce engines like Hadoop. On top of that, the Business Intelligence (BI) field is filled with many other tools that are very specific and likely better-suited than MongoDB.
What is an example of data that is "predistilled or aggregated in
runtime"?
Example of this can be any report that require data from multiple collections.
And why isn't MongoDB very good with it?
In document databases you can't make a join and because of this it hard to build reports. Usually reports it's data aggregating from many tables/collections.
And since mongodb (and document database in general) good fit for data distribution and denormalization better to prebuild reports whenever it possible and just display data from this collection in runtime.
For some tasks/reports it is not possible to prebuild data, in this case mongodb give to you map/reduce, grouping, etc.