The company I work for is running a C# project that crawling data from around 100 websites, saving it to the DB and running some procedures and calculations on that data.
Each one of those 100 websites is having around 10,000 events, and each event is saved to the DB.
After that, the data that was saved is being generated and aggregated to 1 big xml file, so each one of those 10,000 events that were saved, is now presented as a XML file in the DB.
This design looks like that:
1) crawling 100 websites to collects the data and save it the DB.
2) collect the data that was saved to the DB and generate XML files for each event
3) XML files are saved to the DB
The main issue for this post, is the selection of the saved XML files.
Each XML is about 1MB, and considering the fact that there are around 10,000 events, I am not sure SQL Server 2008 R2 is the right option.
I tried to use Redis, and the save is working very well (and fast!), but the query to get those XMLs works very slow (even locally, so network traffic wont be an issue).
I was wondering what are your thoughts? please take into consideration that it is a real-time system, so caching is not an option here.
Any idea will be welcomed.
Thanks.
Instead of using DB you could try a cloud-base system (Azure blobs or Amazon S3), it seems to be a perfect solution. See this post: azure blob storage effectiveness, same situation, except you have XML files instead of images. You can use a DB for storing the metadata, i.e. source and event type of the XML, the path in the cloud, but not the data itself.
You may also zip the files. I don't know the exact method, but it can surely be handled on client-side. Static data is often sent in zipped format to the client by default.
Your question is missing some details such as how long does your data need to remain in the database and such…
I’d avoid storing XML in database if you already have the raw data. Why not have an application that will query the database and generate XML reports on demand? This will save you a lot of space.
10GBs of data per day is something SQL Server 2008 R2 can handle with the right hardware and good structure optimization. You’ll need to investigate if standard edition will be enough or you’ll have to use enterprise or data center licenses.
In any case answer is yes – SQL Server is capable of handling this amount of data but I’d check other solutions as well to see if it’s possible to reduce the costs in any way.
Your basic arch doesn't seem to be at fault, its the way you've perceived the redis, basically if you design your key=>value right there is no way that the retrieval from redis could be slow.
for ex- lets say I have to store 1 mil objects in redis, and say there is an id against which I am storing my objects, this key is nothing but a guid, the save will be really quick, but when it comes to retrieval, do I know the "key" if i KNOW the key it'll be fast, but if I don't know it or I am trying to retrieve my data not on the basis of key but on the basis of some Value in my objects, then off course it'll be slow.
The point is - when it comes to retrieval you should just work against the "Key" and nothing else, so design your key like a pre-calculated value in itself; so when I need to get some data from redis/memcahce, I could make the KEY, and just do a single hit to get the data.
If you could put more details, we'll be able to help you better.
Related
When working with ASP.NET MVC and SQL Server we are wondering if caching to XML is still something to think about or are their other possibilities for this?
Like for instance we have a table called Customers. If you call this db table everytime you click on Customers or do sorting or filtering in the app why not store this info in a xml file.
Then you work only with the xml file and not the db and you update the xml after adding changes to the customers table.
It is an absolutely brilliant idea.
If:
You only have 1 client
Or you have multiple client but they don't mind seeing old data
You have a database system that doesn't provide caching possibilities
You do not use database access frameworks that can handle caching for you
In short, no, it actually is almost never a good idea.
Databases are made to be used. Most of them can handle a much higher load than programmers think they can, as long as you treat them well. If necessary, a lot of them provide perfectly fine caching possibilities to improve performance if needed.
Any useful type of caching in your application should involving refreshing that cache when anything changes. Implementing that by yourself is usually not a good idea. If you do want a very simple cache of data that was just on the screen before the user clicked away, memory would be the place for it, not a file system. Unless you need centralised session cache, but that goes way beyond "let's write some xml".
Caching to xml file is bad choice. Database system can handle load of 100 users in 5 seconds if you have 50000 records in your table. If you want more speed than this then try using In-memory sql which stores data in RAM for fast access. But for it you need high RAM capacity on server.
We are designing an update to a current system (C++\CLI and C#).
The system will gather small (~1Mb) amounts of data from ~10K devices (in the near future). Currently, they are used to save device data in a CSV (a table) and store all these in a wide folder structure.
Data is only inserted (create / append to a file, create folder) never updated / removed.
Data processing is done by reading many CSV's to an external program (like Matlab). Mainly be used for statistical analysis.
There is an option to start saving this data to an MS-SQL database.
Process time (reading the CSV's to external program) could be up to a few minutes.
How should we choose which method to use?
Does one of the methods take significantly more storage than the other?
Roughly, when does reading the raw data from a database becomes quicker than reading the CSV's? (10 files, 100 files? ...)
I'd appreciate your answers, Pros and Cons are welcome.
Thank you for your time.
Well if you are using data in one CSV to get data in another CSV I would guess that SQL Server is going to be faster than whatever you have come up with. I suspect SQL Server would be faster in most cases, but I can't say for sure. Microsoft has put a lot of resources into make a DBMS that does exactly what you are trying to do.
Based on your description it sounds like you have almost created your own DBMS based on table data and folder structure. I suspect that if you switched to using SQL Server you would probably find a number of areas where things are faster and easier.
Possible Pros:
Faster access
Easier to manage
Easier to expand should you need to
Easier to enforce data integrity
Easier to design more complex relationships
Possible Cons:
You would have to rewrite your existing code to use SQL Server instead of your current system
You may have to pay for SQL Server, you would have to check to see if you can use Express
Good luck!
I'd like to try hitting those questions a bit out of order.
Roughly, when does reading the raw data from a database becomes
quicker than reading the CSV's? (10 files, 100 files? ...)
Immediately. The database is optimized (assuming you've done your homework) to read data out at incredible rates.
Does one of the methods take significantly more storage than the
other?
Until you're up in the tens of thousands of files, it probably won't make too much of a difference. Space is cheap, right? However, once you get into the big leagues, you'll notice that the DB is taking up much, much less space.
How should we choose which method to use?
Great question. Everything in the database always comes back to scalability. If you had only a single CSV file to read, you'd be good to go. No DB required. Even dozens, no problem.
It looks like you could end up in a position where you scale up to levels where you'll definitely want the DB engine behind your data pretty quickly. When in doubt, creating a database is the safe bet, since you'll still be able to query that 100 GB worth of data in a second.
This is a question many of our customers have where I work. Unless you need flat files for an existing infrastructure, or you just don't think you can figure out SQL Server, or if you will only have a few files with small amounts of data to manage, you will be better off with SQL Server.
If you have the option to use a ms-sql database, I would do that.
Maintaining data in a wide folder structure is never a good idea. Reading your data would involve reading several files. These could be stored anywhere on your disk. Your file-io time would be quite high. SQL server being a production database has these problems already taken care of.
You are reinventing the wheel here. This is how foxpro manages data, one file per table. It is usually a good idea to use proven technology unless you are actually making a database server.
I do not have any test statistics here, but reading several files will almost always be slower than a database if you are dealing with any significant amount of data. Given your about 10k devices, you should consider using a standard database.
I want to save a whole MS SQL 2008 Database into XML files... using asp.net.
Now I am bit lost here.. what would be the best method to achieve this? Datasets?
And I need to restore the database later again.. using these XML files. I am thinking about using datasets for reading the tables and writing to xml and using the SQLBulkCopy class to restore the database again. But I am not sure whether this would be the right approach..
Any clues and tips for me?
If you will need to restore it on the same server type (I mean SQL Server 2008 or higher) and don't care about ability to see actual data inside the XML do the following:
Programmatically backup the DB using "BACKUP DATABASE" T-SQL
Compress the backup
Convert the backup to Base64
Place the backup as the content of the XML file (like: <database name="..." compressionmethod="..." compressionlevel="...">the Base64 content here</database>
On the server where you need to restore it, download the XML, extract the Base64 content, use the attributes to know what compression was used. Decompress and restore using T-SQL "RESTORE" command.
Would that approach work?
For sure, if you need to see the content of the database, you would need to develop the XML scheme, go through each table etc. But, you won't have SPs/Views and other items backed up.
Because you are talking about a CMS, I'm going to assume you are deploying into hosted environments where you might not have command line access.
Now, before I give you the link I want to state that this is a BAD idea. XML is way too verbose to transfer large amounts of data. Further, although it is relatively easy to pull data out, putting it back in will be difficult and a very time consuming development project in itself.
Next alert: as Denis suggested, you are going to miss all of your stored procedures, functions, etc. Your best bet is to use the normal sql server backup / restore process. (Incidentally, I upvoted his answer).
Finally, the last time I dealt with XML and SQL Server we noticed interesting issues that cropped up when data exceeded a 64KB boundary. Basically, at 63.5KB, the queries ran very quickly (200ms). At 64KB, the query times jumped to over a minute and sometimes quite a bit longer. We didn't bother testing anything over 100KB as that was taking 5 minutes on a fast/dedicated server with zero load.
http://msdn.microsoft.com/en-us/library/ms188273.aspx
See this for putting it back in:
How to insert FOR AUTO XML result into table?
For kicks, here is a link talking about pulling the data out as json objects: http://weblogs.asp.net/thiagosantos/archive/2008/11/17/get-json-from-sql-server.aspx
you should also read (not for the faint of heart): http://www.simple-talk.com/sql/t-sql-programming/consuming-json-strings-in-sql-server/
Of course, the commentors all recommend building something using a CLR approach, but that's probably not available to you in a shared database hosting environment.
At the end of the day, if you are truly insistent on this madness, you might be better served by simply iterating through your table list and exporting all the data to standard CSV files. Then, iterating the CSV files to load the data back in ala C# - is there a way to stream a csv file into database?
Bear in mind that ALL of the above methods suffer from
long processing times due to the data overhead; which leads to
a high potential for failure due to the various time outs (page processing, command, connection, etc); and,
if your data model changes between the time it was exported and reimported then you're back to writing custom translation code and ultimately screwed anyway.
So, only do this if you really really have to and are at least somewhat of a masochist at heart. If the purpose is simply to transfer some data from one installation to another, you might consider using one of the tools like SQL Compare and SQL Data Compare from RedGate to handle the transfer.
I don't care how much (or little) you make, the $1500 investment in their developer bundle is much cheaper than the months of time you are going to spend doing this, fixing it, redoing it, fixing it again, etc. (for the record I do NOT work for them. Their products are just top notch.)
Red Gate's SQL Packager lets you package a database into an exe or to a VS project, so you might want to take a look at that. You can specify which tables you want to consider for data.
Is there any specific reason you want to do this using xml?
I am just beginning to write an application. Part of what it needs to do is to run queries on a database of nutritional information. What I have is the USDA's SR21 Datasets in the form of flat delimited ASCII files.
What I need is advice. I am looking for the best way to import this data into the app and have it easily and quickly queryable at run time. I'll be using it for all the standard things. Populating controls dynamically, Datagrids, calculations, etc. I will also need to do user specific persistent data storage as well. This will not be a commercial app, so hopefully that opens up the possibilities. I am fine with .Net Framework 3.5 so Linq is a possibility when accessing the data (just don't know if it would be the best solution or not). So, what are some suggestions for persistent storage in this scenario? What sort of gotchas should I be watching for? Links to examples are always appreciated of course.
It looks pretty small, so I'd work out an appropriate object model, load the whole lot into memory, and then use LINQ to Objects.
I'm not quite sure what you're asking about in terms of "persistent storage" - aren't you just reading the data? Don't you already have that in the text files? I'm not sure why you'd want to introduce anything else.
I would import the flat files into SQL Server and access via standard ADO.NET functionality. Not only is DB access always better (more robust and powerful) than file I/O as far as data querying and manipulation goes, but you can also take advantage of SQL Server's caching capabilities, especially since this nutritional data won't be changing too often.
If you need to download updated flat files periodically, then look into developing a service that polls for these files and imports into SQL Server automatically.
EDIT: I refer to SQL Server, but feel free to use any DBMS.
My temptation would be to import the data into SQL Server (Express if you aren't looking to deploy the app) as it's a familiar source for me. Alternatively you can probably create an ODBC data source using the text file handler to get you a database-like connection.
I agree that you would benefit from a database, especially for rapid querying, and even more so if you are saving user changes to the data. In order to load the flat file data into a SQL Server (including Express), you can use SSIS.
Use Linq or text data to list method
1.create a list.
2.Read the text file line by line (or all lines).
3.process the line - get required data and attach to the list.
4.process the list for any further use.
the persistence storage will be files and List is volatile.
I'm working on a database driven web-application (ASP.NET, SQL 2008), which receives structured XML data from various sources. The data resembles a set, and often needs 'cleanup', so it is passed through the database as XML, and turned into a resultset for display.
I'd like to capture the produced 'clean' results, and send them to an archive database to persist them to disk.
The options I've considered so far are:
Serialize the entire 'clean' result set into an object (XML/.NET serialized), and send this back to the archive database
PRO: Easily repeatable - can profile/capture the database calls on the archive machine, and re-run them to identify any problems
CON: Versioning could be tricky
Store the cleaned results in a table, and periodically copy fresh records in this table to the archive machine
PRO: Easy build - quick scheduled job
CON: Harder to repro calls on the archive machine; would need to keep input table contents around
Are there any other options, and has anyone had any experience with similar situations?
I have used both cases succesfuly and what I do depends on the system.
Saving Raw Xml:
I tend to save Raw Xml when I am either dealing with unstructured data, or when we are dealing with a messaging system, and we want to track the messages. For example, an application I worked on collected messages from deployed windows clients, we would dump the messages into a relational structure and then roll them into a warehouse. When I took over the project, we started storing the raw xml that was coming because it did allow us the replayability, and the ability to see exactly what is coming into the system.
Relational Data
If I am going to need to do any reporting aggregation of the data, I would break the data out and store it into regular tables. I know you can query xml data in a database, but I try and avoid that. I might still save the original raw messages for replayability and troubleshooting.
Saving a Binary Object
The last thing I have done is save an entire serialized binary object. I find this handy when the object graph is quite complex, and the relationships between the objects are important. It does have a huge downside which is versioning; however, I have managed this quite sucesfully versioning even with namespace changes, object heirarchy changes etc. If you need to access the data in SQL this is not the way to go.