Project title: Implementing push-based query executor

Project Description
Currently, PostgreSQL uses traditional Volcano-style [1] query execution model.
While it is a simple and flexible model, it behaves poorly on modern superscalar
CPUs [2][3] due to lack of locality and frequent instruction mispredictions.
It becomes a major issue for complex OLAP queries with CPU-heavy workloads.
We propose to implement so-called push-based query executor model
as described in [4][5], which improves code and data locality and cache usage
itself; also push-based executor can serve as a platform
for efficient JIT query compilation.

See [6] for more details.

Skills needed
  The ability to understand and modify PostgresSQL executor code;
  The ability to run careful in-memory benchmarks to demonstrate the result;
  The ability to profile Postgres in order to find slow code;
  Understanding modern processors features (pipelining, superscalar CPUs,
    branch prediction, etc) would be very helpful.

Difficulty Level
Moderate-level; however, microoptimizations might be hard.
Probably it will also be hard to keep the whole architecture as clean as it is
now.

Expected Outcomes
Patch with implemented push-based query executor; small patches,
which will optimize the current query executor; benchmarks showing the
performance of queries on plain Postgres and with this patch applied.

References
[1]  Graefe G.. Volcano — an extensible and parallel query evaluation system. IEEE Trans. Knowl. Data Eng.,6(1): 120–135, 1994.
[2] MonetDB/X100: Hyper-Pipelining Query Execution
    http://cidrdb.org/cidr2005/papers/P19.pdf
[3] Vectorization vs. Compilation in Query Execution,
    https://pdfs.semanticscholar.org/dcee/b1e11d3b078b0157325872a581b51402ff66.pdf
[4] Efficiently Compiling Efficient Query Plans for Modern Hardware,
    http://www.vldb.org/pvldb/vol4/p539-neumann.pdf
[5] Compiling Database Queries into Machine Code,
    http://sites.computer.org/debull/A14mar/p3.pdf
[6] [message link, not the whole thread] https://www.postgresql.org/message-id/flat/CAJEAwVFnYMenEe2A9srVuNVemAoW%2BtT_uEs%3D2p427KfsegsJPw%40mail.gmail.com#CAJEAwVFnYMenEe2A9srVuNVemAoW+tT_uEs=2p427KfsegsJPw@mail.gmail.com