NEXMark Benchmark

NEXMark (Niagara Extension to XMark ) is a benchmark we are currently developing in order to benchmark queries over continuous data streams. An initial draft of the benchmark (v0.5) is here. NEXMark is a work in progress and feedback is welcomed. The current schema (for version 0.8, which is not ready yet) is as follows:

Streams:

Person(id, name, emailaddress, creditcard, city, state)
Auction(id, itemname, description, initialbid, reserve, expires, seller, category)
Bid(auction, bidder, price, datetime)

Stored Relations:

Category(id, name, description, parentcategory)

The queries are as follows (in our interpretation of CQL):

Query 1: Currency Conversion

SELECT Istream(auction, DOLTOEUR(price), bidder, datetime)
FROM bid [ROWS UNBOUNDED];

Query 2: Selection

SELECT Rstream(auction, price)
FROM Bid [NOW]
WHERE auction = 1007 OR auction = 1020 OR auction = 2001 OR auction = 2019 OR auction = 2087;

Query 3: Local Item Suggestion

SELECT Istream(P.name, P.city, P.state, A.id)
FROM Auction A [ROWS UNBOUNDED], Person P [ROWS UNBOUNDED]
WHERE A.seller = P.id AND (P.state = `OR' OR P.state = `ID' OR P.state = `CA') AND A.category = 10;

Query 4: Average Price for a Category

SELECT Istream(AVG(Q.final))
FROM Category C, (SELECT Rstream(MAX(B.price) AS final, A.category)
                  FROM Auction A [ROWS UNBOUNDED], Bid B [ROWS UNBOUNDED]
                  WHERE A.id=B.auction AND B.datetime < A.expires AND A.expires < CURRENT_TIME
                  GROUP BY A.id, A.category) Q
WHERE Q.category = C.id
GROUP BY C.id;

Query 5: Hot Items

SELECT Rstream(auction)
FROM (SELECT B1.auction, count(*) AS num
      FROM Bid [RANGE 60 MINUTE SLIDE 1 MINUTE] B1
      GROUP BY B1.auction)
WHERE num >= ALL (SELECT count(*)
                  FROM Bid [RANGE 60 MINUTE SLIDE 1 MINUTE] B2
GROUP BY B2.auction);

Query 6: Average Selling Price by Seller

SELECT Istream(AVG(Q.final), Q.seller)
FROM (SELECT Rstream(MAX(B.price) AS final, A.seller)
      FROM Auction A [ROWS UNBOUNDED], Bid B [ROWS UNBOUNDED]
      WHERE A.id=B.auction AND B.datetime < A.expires AND A.expires < CURRENT_TIME
      GROUP BY A.id, A.seller) [PARTITION BY A.seller ROWS 10] Q
GROUP BY Q.seller;

Query 7: Highest Bid

SELECT Rstream(B.auction, B.price, B.bidder)
FROM Bid [RANGE 1 MINUTE SLIDE 1 MINUTE] B
WHERE B.price = (SELECT MAX(B1.price)
                 FROM BID [RANGE 1 MINUTE SLIDE 1 MINUTE] B1);

Query 8: Monitor New Users

SELECT Rstream(P.id, P.name, A.reserve)
FROM Person [RANGE 12 HOUR] P, Auction [RANGE 12 HOUR] A
WHERE P.id = A.seller;

The NEXMarkGenerator (in java) is here.