WIP: NOT READY FOR REVIEW: Add parquet progressive_io option to keep one-shot row group I/O with pushdown filters

[alamb]

Which issue does this PR close?

• Part of Parquet Pushdown Filters: Way to keep the I/O pattern the same when pushdown filters are enabled #24393
• Related to [EPIC] Fix performance regressions when enabling parquet filter pushdown (late materialization) #20324 and Enable parquet filter pushdown (filter_pushdown) by default #3463

Note: this is a prototype and is NOT READY FOR REVIEW. It exists to evaluate whether decoupling the I/O pattern from filter pushdown avoids the I/O-related regressions that have blocked enabling pushdown_filters by default.

Rationale for this change

Enabling pushdown_filters today also changes the I/O pattern: instead of one request per row group, the reader sequentially fetches the columns for each filter predicate, then (after evaluating the filters) the remaining projected columns. When a file has no offset index (e.g. the ClickBench dataset), those extra round trips cannot prune any bytes, so they are pure overhead.

This PR separates "filter evaluation mechanics" from the I/O pattern via a new config option, datafusion.execution.parquet.progressive_io:

pushdown_filters	progressive_io	behavior
false	(any)	unchanged: one request per row group
true	false (default)	filters evaluated during decode, but all column chunks needed by the filters or the projection are fetched with one request per row group
true	true	progressive fetch, but only when the file has an offset index (where it can actually prune bytes); otherwise one request per row group

What changes are included in this PR?

• New progressive_io config option (config, proto, docs, ParquetSource::with_progressive_io), resolved table-or-session like pushdown_filters
• RowFilterGenerator now exposes the union of the filter predicates' ProjectionMasks
• In the opener, when one-shot I/O applies, the byte ranges of all column chunks in (filter mask ∪ projection mask) are precomputed per row group; PushDecoderStreamState::transition expands the first NeedsData request of each row group to those ranges, so the whole row group is fetched in one get_byte_ranges call. The decoder's buffered-range containment check then satisfies all subsequent phases with zero additional I/O. No arrow-rs changes were needed.

Known prototype limitations:

• When page-index pruning produces a row selection that actually prunes rows, the scan falls back to progressive I/O (one-shot whole-chunk ranges would defeat the page pruning). A future refinement could fetch selection-pruned page ranges in a single request instead.
• One-shot ranges are whole column chunks, so peak buffered bytes per row group equal the compressed size of all needed chunks (same as the non-pushdown reader today).

Are these changes tested?

Yes:

• New object-store request-count tests in object_store_access.rs show that with pushdown_filters=true and the default progressive_io=false, the scan issues exactly one data request per row group with byte ranges identical to the non-pushdown scan, and that progressive_io=true restores the current progressive pattern
• Existing filter pushdown tests and sqllogictests pass unchanged

Are there any user-facing changes?

Yes: a new configuration option, and (intentionally) a changed default I/O pattern for users who enable pushdown_filters. Benchmarks (clickbench_pushdown) to follow.

[alamb]

run benchmark clickbench_partitioned

[alamb]

run benchmark clickbench_pushdown

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5307167906-1616-gtt5l 6.12.85+ #1 SMP Wed Jun 17 20:31:55 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing alamb/progressive_io (4342770) to 00eba79 (merge-base) diff

Run configuration

run benchmark clickbench_partitioned

Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5307170440-1617-45h8f 6.12.85+ #1 SMP Wed Jun 17 20:31:55 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing alamb/progressive_io (4342770) to 00eba79 (merge-base) diff

Run configuration

run benchmark clickbench_pushdown

Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing alamb/progressive_io (4342770) to 00eba79 (merge-base) diff

Run configuration

run benchmark clickbench_pushdown

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and alamb_progressive_io
--------------------
Benchmark clickbench_pushdown.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ alamb_progressive_io ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │    1.23 ms │              1.24 ms │     no change │
│ QQuery 1  │   12.05 ms │             12.02 ms │     no change │
│ QQuery 2  │   35.87 ms │             36.41 ms │     no change │
│ QQuery 3  │   31.04 ms │             31.25 ms │     no change │
│ QQuery 4  │  222.36 ms │            223.81 ms │     no change │
│ QQuery 5  │  273.29 ms │            271.64 ms │     no change │
│ QQuery 6  │    1.30 ms │              1.27 ms │     no change │
│ QQuery 7  │   15.55 ms │             15.79 ms │     no change │
│ QQuery 8  │  325.73 ms │            320.85 ms │     no change │
│ QQuery 9  │  447.39 ms │            445.43 ms │     no change │
│ QQuery 10 │   92.47 ms │             92.81 ms │     no change │
│ QQuery 11 │  103.16 ms │            103.58 ms │     no change │
│ QQuery 12 │  291.74 ms │            296.16 ms │     no change │
│ QQuery 13 │  409.75 ms │            392.67 ms │     no change │
│ QQuery 14 │  311.60 ms │            310.33 ms │     no change │
│ QQuery 15 │  271.61 ms │            267.29 ms │     no change │
│ QQuery 16 │  607.02 ms │            612.19 ms │     no change │
│ QQuery 17 │  618.56 ms │            614.36 ms │     no change │
│ QQuery 18 │ 1240.28 ms │           1261.16 ms │     no change │
│ QQuery 19 │   28.51 ms │             28.60 ms │     no change │
│ QQuery 20 │  514.57 ms │            516.16 ms │     no change │
│ QQuery 21 │  554.74 ms │            559.65 ms │     no change │
│ QQuery 22 │  904.31 ms │            915.27 ms │     no change │
│ QQuery 23 │  114.11 ms │            112.69 ms │     no change │
│ QQuery 24 │   36.72 ms │             34.12 ms │ +1.08x faster │
│ QQuery 25 │  138.42 ms │            141.03 ms │     no change │
│ QQuery 26 │   48.61 ms │             48.46 ms │     no change │
│ QQuery 27 │  568.29 ms │            552.57 ms │     no change │
│ QQuery 28 │ 2905.12 ms │           2923.80 ms │     no change │
│ QQuery 29 │   40.66 ms │             41.17 ms │     no change │
│ QQuery 30 │  301.78 ms │            312.71 ms │     no change │
│ QQuery 31 │  278.73 ms │            284.82 ms │     no change │
│ QQuery 32 │  929.57 ms │            940.37 ms │     no change │
│ QQuery 33 │ 1428.21 ms │           1433.23 ms │     no change │
│ QQuery 34 │ 1464.64 ms │           1463.34 ms │     no change │
│ QQuery 35 │  279.84 ms │            280.79 ms │     no change │
│ QQuery 36 │   68.08 ms │             68.32 ms │     no change │
│ QQuery 37 │   38.05 ms │             35.62 ms │ +1.07x faster │
│ QQuery 38 │   34.77 ms │             35.80 ms │     no change │
│ QQuery 39 │  135.44 ms │            126.98 ms │ +1.07x faster │
│ QQuery 40 │   17.02 ms │             17.00 ms │     no change │
│ QQuery 41 │   16.15 ms │             15.89 ms │     no change │
│ QQuery 42 │   13.96 ms │             13.94 ms │     no change │
└───────────┴────────────┴──────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                   ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                   │ 16172.32ms │
│ Total Time (alamb_progressive_io)   │ 16212.58ms │
│ Average Time (HEAD)                 │   376.10ms │
│ Average Time (alamb_progressive_io) │   377.04ms │
│ Queries Faster                      │          3 │
│ Queries Slower                      │          0 │
│ Queries with No Change              │         40 │
│ Queries with Failure                │          0 │
└─────────────────────────────────────┴────────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and alamb_progressive_io
--------------------
Benchmark clickbench_pushdown.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃                  alamb_progressive_io ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.23 / 3.88 ±5.25 / 14.37 ms │          1.24 / 3.90 ±5.26 / 14.42 ms │     no change │
│ QQuery 1  │        12.05 / 12.26 ±0.12 / 12.41 ms │        12.02 / 12.29 ±0.16 / 12.44 ms │     no change │
│ QQuery 2  │        35.87 / 36.25 ±0.24 / 36.49 ms │        36.41 / 36.95 ±0.38 / 37.48 ms │     no change │
│ QQuery 3  │        31.04 / 31.65 ±0.69 / 32.95 ms │        31.25 / 31.36 ±0.14 / 31.62 ms │     no change │
│ QQuery 4  │     222.36 / 225.27 ±2.42 / 228.70 ms │     223.81 / 224.93 ±1.04 / 226.42 ms │     no change │
│ QQuery 5  │     273.29 / 274.61 ±0.96 / 276.09 ms │     271.64 / 273.27 ±1.56 / 275.50 ms │     no change │
│ QQuery 6  │           1.30 / 1.44 ±0.22 / 1.86 ms │           1.27 / 1.41 ±0.20 / 1.81 ms │     no change │
│ QQuery 7  │        15.55 / 15.76 ±0.15 / 15.99 ms │        15.79 / 15.97 ±0.12 / 16.14 ms │     no change │
│ QQuery 8  │     325.73 / 329.97 ±2.62 / 332.82 ms │     320.85 / 326.09 ±2.80 / 329.26 ms │     no change │
│ QQuery 9  │     447.39 / 454.93 ±4.53 / 460.15 ms │     445.43 / 456.64 ±8.04 / 469.64 ms │     no change │
│ QQuery 10 │        92.47 / 93.06 ±0.80 / 94.64 ms │        92.81 / 93.46 ±0.70 / 94.64 ms │     no change │
│ QQuery 11 │     103.16 / 104.20 ±0.67 / 105.03 ms │     103.58 / 105.01 ±1.71 / 108.01 ms │     no change │
│ QQuery 12 │     291.74 / 300.13 ±6.32 / 305.52 ms │     296.16 / 299.40 ±3.94 / 307.05 ms │     no change │
│ QQuery 13 │     409.75 / 424.08 ±7.21 / 429.13 ms │     392.67 / 404.47 ±8.50 / 416.90 ms │     no change │
│ QQuery 14 │     311.60 / 314.76 ±3.43 / 319.14 ms │     310.33 / 317.32 ±6.94 / 329.18 ms │     no change │
│ QQuery 15 │     271.61 / 274.36 ±2.58 / 278.97 ms │    267.29 / 280.77 ±15.42 / 310.36 ms │     no change │
│ QQuery 16 │     607.02 / 616.79 ±6.03 / 624.37 ms │    612.19 / 623.43 ±10.92 / 644.20 ms │     no change │
│ QQuery 17 │     618.56 / 626.61 ±5.86 / 632.90 ms │     614.36 / 624.54 ±6.41 / 631.56 ms │     no change │
│ QQuery 18 │ 1240.28 / 1257.90 ±14.01 / 1278.61 ms │ 1261.16 / 1277.54 ±16.95 / 1302.90 ms │     no change │
│ QQuery 19 │        28.51 / 35.21 ±8.17 / 50.69 ms │       28.60 / 36.39 ±10.31 / 54.44 ms │     no change │
│ QQuery 20 │    514.57 / 522.95 ±11.45 / 545.58 ms │     516.16 / 521.96 ±6.02 / 532.60 ms │     no change │
│ QQuery 21 │    554.74 / 574.41 ±11.29 / 587.32 ms │     559.65 / 573.51 ±9.11 / 582.61 ms │     no change │
│ QQuery 22 │     904.31 / 912.65 ±7.48 / 925.47 ms │     915.27 / 925.54 ±6.39 / 934.13 ms │     no change │
│ QQuery 23 │     114.11 / 116.02 ±1.22 / 117.98 ms │    112.69 / 128.05 ±12.91 / 146.25 ms │  1.10x slower │
│ QQuery 24 │        36.72 / 43.00 ±6.33 / 51.18 ms │        34.12 / 38.96 ±7.71 / 54.35 ms │ +1.10x faster │
│ QQuery 25 │    138.42 / 145.60 ±11.43 / 168.35 ms │     141.03 / 143.09 ±1.76 / 145.77 ms │     no change │
│ QQuery 26 │        48.61 / 51.22 ±3.02 / 56.65 ms │        48.46 / 49.93 ±2.03 / 53.68 ms │     no change │
│ QQuery 27 │     568.29 / 575.36 ±6.49 / 583.91 ms │    552.57 / 568.44 ±10.98 / 582.70 ms │     no change │
│ QQuery 28 │ 2905.12 / 2916.99 ±10.62 / 2935.64 ms │ 2923.80 / 2940.99 ±16.93 / 2971.92 ms │     no change │
│ QQuery 29 │        40.66 / 44.31 ±6.88 / 58.07 ms │        41.17 / 44.87 ±6.45 / 57.72 ms │     no change │
│ QQuery 30 │     301.78 / 311.24 ±7.64 / 321.89 ms │    312.71 / 323.19 ±12.56 / 343.76 ms │     no change │
│ QQuery 31 │     278.73 / 293.33 ±8.90 / 306.27 ms │     284.82 / 291.09 ±3.79 / 295.07 ms │     no change │
│ QQuery 32 │    929.57 / 957.16 ±24.10 / 986.73 ms │   940.37 / 970.79 ±27.50 / 1021.68 ms │     no change │
│ QQuery 33 │ 1428.21 / 1459.93 ±25.27 / 1490.99 ms │ 1433.23 / 1475.20 ±22.60 / 1500.43 ms │     no change │
│ QQuery 34 │ 1464.64 / 1510.69 ±29.67 / 1545.88 ms │ 1463.34 / 1488.49 ±17.60 / 1511.26 ms │     no change │
│ QQuery 35 │    279.84 / 309.37 ±34.26 / 365.66 ms │    280.79 / 298.94 ±19.39 / 333.63 ms │     no change │
│ QQuery 36 │        68.08 / 72.07 ±2.45 / 75.40 ms │       68.32 / 81.98 ±10.58 / 92.10 ms │  1.14x slower │
│ QQuery 37 │        38.05 / 43.51 ±5.78 / 54.25 ms │        35.62 / 38.24 ±3.00 / 43.90 ms │ +1.14x faster │
│ QQuery 38 │        34.77 / 37.53 ±2.58 / 41.48 ms │        35.80 / 39.51 ±4.45 / 47.47 ms │  1.05x slower │
│ QQuery 39 │    135.44 / 149.79 ±10.54 / 167.11 ms │     126.98 / 143.59 ±8.40 / 149.32 ms │     no change │
│ QQuery 40 │        17.02 / 17.54 ±0.46 / 18.23 ms │        17.00 / 17.84 ±1.39 / 20.59 ms │     no change │
│ QQuery 41 │        16.15 / 16.29 ±0.15 / 16.59 ms │        15.89 / 16.24 ±0.33 / 16.77 ms │     no change │
│ QQuery 42 │        13.96 / 15.76 ±3.13 / 22.00 ms │        13.94 / 15.70 ±3.14 / 21.97 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                   ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                   │ 16529.85ms │
│ Total Time (alamb_progressive_io)   │ 16581.30ms │
│ Average Time (HEAD)                 │   384.42ms │
│ Average Time (alamb_progressive_io) │   385.61ms │
│ Queries Faster                      │          2 │
│ Queries Slower                      │          3 │
│ Queries with No Change              │         38 │
│ Queries with Failure                │          0 │
└─────────────────────────────────────┴────────────┘

Resource Usage

clickbench_pushdown — base (merge-base)

Metric	Value
Wall time	85.0s
Peak memory	11.9 GiB
Avg memory	4.8 GiB
CPU user	839.8s
CPU sys	58.8s
Peak spill	0 B

clickbench_pushdown — branch

Metric	Value
Wall time	85.0s
Peak memory	12.4 GiB
Avg memory	5.2 GiB
CPU user	841.9s
CPU sys	61.8s
Peak spill	0 B

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing alamb/progressive_io (4342770) to 00eba79 (merge-base) diff

Run configuration

run benchmark clickbench_partitioned

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and alamb_progressive_io
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ alamb_progressive_io ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │    1.30 ms │              1.28 ms │     no change │
│ QQuery 1  │   12.11 ms │             12.63 ms │     no change │
│ QQuery 2  │   36.64 ms │             37.93 ms │     no change │
│ QQuery 3  │   32.12 ms │             32.58 ms │     no change │
│ QQuery 4  │  245.81 ms │            249.16 ms │     no change │
│ QQuery 5  │  292.41 ms │            294.88 ms │     no change │
│ QQuery 6  │    1.33 ms │              1.37 ms │     no change │
│ QQuery 7  │   13.72 ms │             14.00 ms │     no change │
│ QQuery 8  │  358.00 ms │            360.07 ms │     no change │
│ QQuery 9  │  490.53 ms │            518.93 ms │  1.06x slower │
│ QQuery 10 │   72.28 ms │             74.79 ms │     no change │
│ QQuery 11 │   83.53 ms │             85.47 ms │     no change │
│ QQuery 12 │  287.55 ms │            296.48 ms │     no change │
│ QQuery 13 │  389.51 ms │            405.15 ms │     no change │
│ QQuery 14 │  304.94 ms │            311.65 ms │     no change │
│ QQuery 15 │  305.15 ms │            310.03 ms │     no change │
│ QQuery 16 │  667.64 ms │            662.01 ms │     no change │
│ QQuery 17 │  666.61 ms │            680.08 ms │     no change │
│ QQuery 18 │ 1355.59 ms │           1390.22 ms │     no change │
│ QQuery 19 │   29.08 ms │             29.36 ms │     no change │
│ QQuery 20 │  529.59 ms │            526.98 ms │     no change │
│ QQuery 21 │  527.02 ms │            528.32 ms │     no change │
│ QQuery 22 │ 1011.03 ms │           1025.95 ms │     no change │
│ QQuery 23 │ 3190.66 ms │           3252.03 ms │     no change │
│ QQuery 24 │   42.27 ms │             42.17 ms │     no change │
│ QQuery 25 │  116.03 ms │            115.76 ms │     no change │
│ QQuery 26 │   42.31 ms │             43.30 ms │     no change │
│ QQuery 27 │  533.13 ms │            530.13 ms │     no change │
│ QQuery 28 │ 2985.77 ms │           3021.03 ms │     no change │
│ QQuery 29 │   42.10 ms │             42.54 ms │     no change │
│ QQuery 30 │  329.55 ms │            328.23 ms │     no change │
│ QQuery 31 │  302.64 ms │            304.70 ms │     no change │
│ QQuery 32 │ 1030.28 ms │            993.71 ms │     no change │
│ QQuery 33 │ 1578.18 ms │           1598.79 ms │     no change │
│ QQuery 34 │ 1608.59 ms │           1557.41 ms │     no change │
│ QQuery 35 │  320.77 ms │            330.93 ms │     no change │
│ QQuery 36 │   75.88 ms │             76.91 ms │     no change │
│ QQuery 37 │   38.12 ms │             37.16 ms │     no change │
│ QQuery 38 │   42.43 ms │             44.28 ms │     no change │
│ QQuery 39 │  160.27 ms │            150.02 ms │ +1.07x faster │
│ QQuery 40 │   15.42 ms │             15.26 ms │     no change │
│ QQuery 41 │   15.00 ms │             15.30 ms │     no change │
│ QQuery 42 │   14.19 ms │             14.43 ms │     no change │
└───────────┴────────────┴──────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                   ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                   │ 20197.03ms │
│ Total Time (alamb_progressive_io)   │ 20363.45ms │
│ Average Time (HEAD)                 │   469.70ms │
│ Average Time (alamb_progressive_io) │   473.57ms │
│ Queries Faster                      │          1 │
│ Queries Slower                      │          1 │
│ Queries with No Change              │         41 │
│ Queries with Failure                │          0 │
└─────────────────────────────────────┴────────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and alamb_progressive_io
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃                  alamb_progressive_io ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.30 / 4.20 ±5.66 / 15.51 ms │          1.28 / 4.21 ±5.70 / 15.61 ms │     no change │
│ QQuery 1  │        12.11 / 12.59 ±0.27 / 12.87 ms │        12.63 / 12.93 ±0.23 / 13.26 ms │     no change │
│ QQuery 2  │        36.64 / 37.17 ±0.33 / 37.59 ms │        37.93 / 38.67 ±0.51 / 39.27 ms │     no change │
│ QQuery 3  │        32.12 / 32.68 ±0.72 / 34.09 ms │        32.58 / 32.86 ±0.23 / 33.14 ms │     no change │
│ QQuery 4  │     245.81 / 250.19 ±2.32 / 252.22 ms │     249.16 / 253.41 ±3.57 / 258.91 ms │     no change │
│ QQuery 5  │     292.41 / 296.00 ±3.67 / 301.12 ms │     294.88 / 300.41 ±4.11 / 305.04 ms │     no change │
│ QQuery 6  │           1.33 / 1.49 ±0.23 / 1.94 ms │           1.37 / 1.52 ±0.23 / 1.98 ms │     no change │
│ QQuery 7  │        13.72 / 14.02 ±0.23 / 14.30 ms │        14.00 / 14.34 ±0.20 / 14.56 ms │     no change │
│ QQuery 8  │     358.00 / 360.50 ±1.38 / 361.82 ms │     360.07 / 364.86 ±4.40 / 372.19 ms │     no change │
│ QQuery 9  │     490.53 / 500.49 ±8.41 / 515.39 ms │    518.93 / 537.20 ±18.31 / 560.44 ms │  1.07x slower │
│ QQuery 10 │        72.28 / 77.03 ±6.57 / 89.99 ms │        74.79 / 75.53 ±0.63 / 76.52 ms │     no change │
│ QQuery 11 │        83.53 / 84.72 ±0.93 / 86.32 ms │        85.47 / 87.72 ±1.91 / 91.17 ms │     no change │
│ QQuery 12 │     287.55 / 295.74 ±7.46 / 308.49 ms │     296.48 / 301.32 ±6.17 / 313.02 ms │     no change │
│ QQuery 13 │    389.51 / 400.26 ±14.33 / 428.24 ms │    405.15 / 417.65 ±11.94 / 432.61 ms │     no change │
│ QQuery 14 │     304.94 / 310.17 ±4.78 / 317.62 ms │     311.65 / 318.73 ±4.44 / 325.10 ms │     no change │
│ QQuery 15 │     305.15 / 311.63 ±8.38 / 327.40 ms │    310.03 / 318.71 ±12.22 / 342.51 ms │     no change │
│ QQuery 16 │     667.64 / 676.86 ±7.07 / 689.19 ms │     662.01 / 671.67 ±7.58 / 685.17 ms │     no change │
│ QQuery 17 │     666.61 / 671.40 ±3.61 / 675.90 ms │    680.08 / 701.55 ±19.08 / 735.42 ms │     no change │
│ QQuery 18 │ 1355.59 / 1386.15 ±24.30 / 1414.98 ms │ 1390.22 / 1428.55 ±23.15 / 1463.01 ms │     no change │
│ QQuery 19 │      29.08 / 45.56 ±31.87 / 109.28 ms │        29.36 / 31.52 ±2.61 / 36.24 ms │ +1.45x faster │
│ QQuery 20 │     529.59 / 539.84 ±9.00 / 551.96 ms │    526.98 / 543.91 ±12.03 / 557.80 ms │     no change │
│ QQuery 21 │     527.02 / 533.91 ±4.52 / 540.19 ms │     528.32 / 537.44 ±6.53 / 548.17 ms │     no change │
│ QQuery 22 │  1011.03 / 1026.48 ±9.54 / 1037.42 ms │  1025.95 / 1032.31 ±5.83 / 1043.33 ms │     no change │
│ QQuery 23 │ 3190.66 / 3233.71 ±25.48 / 3264.66 ms │ 3252.03 / 3286.93 ±25.95 / 3319.69 ms │     no change │
│ QQuery 24 │        42.27 / 42.94 ±0.53 / 43.82 ms │        42.17 / 44.42 ±3.86 / 52.11 ms │     no change │
│ QQuery 25 │    116.03 / 124.01 ±14.98 / 153.95 ms │     115.76 / 125.81 ±8.80 / 137.69 ms │     no change │
│ QQuery 26 │        42.31 / 43.33 ±0.70 / 44.32 ms │        43.30 / 43.87 ±0.88 / 45.62 ms │     no change │
│ QQuery 27 │     533.13 / 537.49 ±5.76 / 548.28 ms │    530.13 / 542.60 ±17.19 / 575.50 ms │     no change │
│ QQuery 28 │ 2985.77 / 2998.65 ±13.99 / 3023.66 ms │ 3021.03 / 3041.54 ±19.31 / 3074.08 ms │     no change │
│ QQuery 29 │        42.10 / 46.03 ±5.69 / 57.06 ms │        42.54 / 52.18 ±7.77 / 59.74 ms │  1.13x slower │
│ QQuery 30 │     329.55 / 332.31 ±2.43 / 336.87 ms │    328.23 / 343.00 ±13.50 / 359.69 ms │     no change │
│ QQuery 31 │     302.64 / 311.36 ±8.73 / 327.07 ms │    304.70 / 321.35 ±12.10 / 337.39 ms │     no change │
│ QQuery 32 │ 1030.28 / 1060.85 ±26.92 / 1095.97 ms │  993.71 / 1020.24 ±17.46 / 1046.30 ms │     no change │
│ QQuery 33 │ 1578.18 / 1602.91 ±24.34 / 1649.47 ms │ 1598.79 / 1619.40 ±17.50 / 1641.05 ms │     no change │
│ QQuery 34 │ 1608.59 / 1623.04 ±13.75 / 1645.60 ms │ 1557.41 / 1610.57 ±29.08 / 1644.31 ms │     no change │
│ QQuery 35 │    320.77 / 362.65 ±34.84 / 410.24 ms │    330.93 / 346.63 ±19.32 / 383.70 ms │     no change │
│ QQuery 36 │        75.88 / 81.76 ±6.21 / 89.44 ms │        76.91 / 82.24 ±5.31 / 91.81 ms │     no change │
│ QQuery 37 │        38.12 / 41.26 ±4.06 / 48.73 ms │        37.16 / 37.92 ±0.60 / 39.01 ms │ +1.09x faster │
│ QQuery 38 │        42.43 / 44.19 ±1.34 / 46.10 ms │        44.28 / 47.50 ±4.87 / 56.95 ms │  1.07x slower │
│ QQuery 39 │     160.27 / 167.95 ±5.47 / 175.32 ms │    150.02 / 167.72 ±12.29 / 181.15 ms │     no change │
│ QQuery 40 │        15.42 / 15.76 ±0.62 / 16.99 ms │        15.26 / 20.73 ±7.32 / 34.07 ms │  1.32x slower │
│ QQuery 41 │        15.00 / 15.18 ±0.16 / 15.40 ms │        15.30 / 18.79 ±2.86 / 21.61 ms │  1.24x slower │
│ QQuery 42 │        14.19 / 14.70 ±0.28 / 15.06 ms │        14.43 / 18.07 ±5.88 / 29.79 ms │  1.23x slower │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                   ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                   │ 20569.17ms │
│ Total Time (alamb_progressive_io)   │ 20818.51ms │
│ Average Time (HEAD)                 │   478.35ms │
│ Average Time (alamb_progressive_io) │   484.15ms │
│ Queries Faster                      │          2 │
│ Queries Slower                      │          6 │
│ Queries with No Change              │         35 │
│ Queries with Failure                │          0 │
└─────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	105.0s
Peak memory	12.4 GiB
Avg memory	4.7 GiB
CPU user	1053.2s
CPU sys	78.5s
Peak spill	0 B

clickbench_partitioned — branch

Metric	Value
Wall time	105.0s
Peak memory	11.3 GiB
Avg memory	4.2 GiB
CPU user	1057.2s
CPU sys	81.2s
Peak spill	0 B

---

File an issue against this benchmark runner