gistfile1.txt

plist(parallel(workerCollection,
               numWorkers="1",
               sort="year asc, month asc, day asc",
               plist(facet(week1, zkHost="zk1", q="*:*", buckets="year, month, day", bucketSorts="year asc, month asc, day asc", count(*)), 
                     facet(week2, zkHost="zk1", q="*:*", buckets="year, month, day", bucketSorts="year asc, month asc, day asc", count(*)),
                     facet(week3, zkHost="zk1", q="*:*", buckets="year, month, day", bucketSorts="year asc, month asc, day asc", count(*)))),
     parallel(workerCollection,
              numWorkers="1",
              sort="year asc, month asc, day asc",
              plist(facet(week4, zkHost="zk2", q="*:*", buckets="year, month, day", bucketSorts="year asc, month asc, day asc", count(*)),
                    facet(week5, zkHost="zk2", q="*:*", buckets="year, month, day", bucketSorts="year asc, month asc, day asc", count(*)),
                    facet(week6, zkHost="zk2", q="*:*", buckets="year, month, day", bucketSorts="year asc, month asc, day asc", count(*)))), 
      parallel(workerCollection,
               numWorkers="1",
               sort="year asc, month asc, day asc",
               plist(facet(week7, zkHost="zk3", q="*:*", buckets="year, month, day", bucketSorts="year asc, month asc, day asc", count(*)),
                     facet(week8, zkHost="zk3", q="*:*", buckets="year, month, day", bucketSorts="year asc, month asc, day asc", count(*)),
                     facet(week9, zkHost="zk3", q="*:*", buckets="year, month, day", bucketSorts="year asc, month asc, day asc", count(*)))))

The data is organized in chronological Solr Cloud collections by week. New collections are added every week. Each collection has two shards and two replicas and 200+ million records. Aggregating across 10 years of data would involve:

104,000,000,000 documents
520 collections
2080 Solr Instances