Generic class for performing mass-parallel queries. Feedback?

Posted by Aaron on Stack Overflow See other posts from Stack Overflow or by Aaron
Published on 2010-12-26T22:53:37Z Indexed on 2010/12/27 2:54 UTC
Read the original article Hit count: 217

I don't understand why, but there appears to be no mechanism in the client library for performing many queries in parallel for Windows Azure Table Storage. I've created a template class that can be used to save considerable time, and you're welcome to use it however you wish. I would appreciate however, if you could pick it apart, and provide feedback on how to improve this class.

public class AsyncDataQuery<T> where T: new()
{
    public AsyncDataQuery(bool preserve_order)
    {
        m_preserve_order = preserve_order;
        this.Queries = new List<CloudTableQuery<T>>(1000);
    }

    public void AddQuery(IQueryable<T> query)
    {
        var data_query = (DataServiceQuery<T>)query;
        var uri = data_query.RequestUri; // required

        this.Queries.Add(new CloudTableQuery<T>(data_query));
    }

    /// <summary>
    /// Blocking but still optimized.
    /// </summary>
    public List<T> Execute()
    {
        this.BeginAsync();
        return this.EndAsync();
    }

    public void BeginAsync()
    {
        if (m_preserve_order == true)
        {
            this.Items = new List<T>(Queries.Count);
            for (var i = 0; i < Queries.Count; i++)
            {
                this.Items.Add(new T());
            }
        }
        else
        {
            this.Items = new List<T>(Queries.Count * 2);
        }

        m_wait = new ManualResetEvent(false);

        for (var i = 0; i < Queries.Count; i++)
        {
            var query = Queries[i];
            query.BeginExecuteSegmented(callback, i);
        }
    }

    public List<T> EndAsync()
    {
        m_wait.WaitOne();

        return this.Items;
    }

    private List<T> Items { get; set; }
    private List<CloudTableQuery<T>> Queries { get; set; }

    private bool m_preserve_order;
    private ManualResetEvent m_wait;
    private int m_completed = 0;

    private void callback(IAsyncResult ar)
    {
        int i = (int)ar.AsyncState;
        CloudTableQuery<T> query = Queries[i];
        var response = query.EndExecuteSegmented(ar);
        if (m_preserve_order == true)
        { // preserve ordering only supports one result per query
            this.Items[i] = response.Results.First();
        }
        else
        { // add any number of items
            this.Items.AddRange(response.Results);
        }
        if (response.HasMoreResults == true)
        { // more data to pull
            query.BeginExecuteSegmented(response.ContinuationToken, callback, i);
            return;
        }
        m_completed = Interlocked.Increment(ref m_completed);
        if (m_completed == Queries.Count)
        {
            m_wait.Set();
        }
    }
}

© Stack Overflow or respective owner

Related posts about c#

Related posts about Azure