compactness.js

// compactness() calculates how closely the resulting documents are located together
// It counts the size of the documents vs size of the unique pages they reside on
 
function compactness(collection, query, limit) {
    Object.size = function(o) { var size = 0, key;
        for (key in o) { if (o.hasOwnProperty(key)) size++; }
        return size; };
 
    count=0;
    size=0;
    blocksize=1024*4;
    disklocs={} //will store each disk loc, format: file-loc%4kb, ie 1-4096, etc
 
    db.getCollection(collection).find(query).limit(limit).showDiskLoc().forEach(
            function(doc) {
                count++;
                docsize=Object.bsonsize(doc)-45 //$diskloc info adds 45 bytes
                size+=docsize 
                file=doc.$diskLoc.file
                offset=(doc.$diskLoc.offset)
                offsetPage=offset-offset%4096
                disklocs[file+"-"+offsetPage]=1;
 
                if (docsize > blocksize) {
                    //if document size is bigger than 4kb, need to count the next few pages as well
                    numPagesPerDoc = Math.ceil(docsize/(blocksize))
                    for(i=0;i<numPagesPerDoc;i++){
                        disklocs[file+"-"+ (offsetPage + blocksize * i) ]=1;
                    }
                }
            }
        )
    numpages=Object.size(disklocs)
    numbytespages=blocksize*numpages;
    printjson(disklocs);
    print("Number of documents returned  : " + count)
    print("Number of pages touched       : " + numpages)
    print("Size of returned data in bytes: " + size)
    print("Size of pages touched by data : " + numbytespages)
    print("Compactness: " + Math.floor(100*size/numbytespages) + "%");
}
 
/* Example usage:
> compactness("large",{},1);
Number of documents returned  : 1
Number of pages touched       : 2
Size of returned data in bytes: 5142
Size of pages touched by data : 8192
Compactness: 62% */