[GraphBolt] Check return values of liburing calls and refactor.

[mfbalin]

Description

In the old code, we were not checking the return value of the results. If the reads are short (requested N bytes but it reads M < N bytes), then we have to continue from where it left off and submit another request. Now, we can write a regression test
and measure speed and after speed is measured, check correctness without affecting timing.

Regression test:

import os
import tempfile
from functools import partial
import time

import torch
import dgl.graphbolt as gb

assert_equal = partial(torch.testing.assert_close, rtol=0, atol=0)

def to_on_disk_numpy(test_dir, name, t):
    path = os.path.join(test_dir, name + ".npy")
    gb.numpy_save_aligned(path, t.numpy())
    return path

def test_index_select_throughput_and_iops(shape, dtype, indices, num_threads_list):
    tensor = torch.randint(0, 127, shape, dtype=dtype)

    skip_first = 10

    results = []
    IOPSs = []

    with tempfile.TemporaryDirectory() as test_dir:
        path = to_on_disk_numpy(test_dir, "tensor", tensor)

        for num_threads in num_threads_list:
            feature = gb.DiskBasedFeature(path=path, num_threads=num_threads)

            throughput_sum = 0
            iops_sum = 0

            for i, idx in enumerate(indices):
                start = time.time()
                result = feature.read(idx)
                duration = time.time() - start
                assert_equal(result, tensor[idx])

                if i >= skip_first:
                    throughput_sum += result.nbytes / duration
                    iops_sum += idx.numel() / duration
    
            throughput = throughput_sum / (len(indices) - skip_first)
            iops = iops_sum / (len(indices) - skip_first)
            print(num_threads, int(throughput / (2 ** 20)), "MiB/s", int(iops), "IOPS")
            results.append(throughput)
            IOPSs.append(iops)

    return results, IOPSs

shape = [2500000, 4096]
dtype = torch.int8
batch_size = 100000
indices = [torch.randint(0, shape[0], [batch_size], dtype=torch.int32) for _ in range(25)]
num_threads_list = list(range(1, 9))

throughputs, IOPSs = test_index_select_throughput_and_iops(shape, dtype, indices, num_threads_list)
print(list((num_threads, int(throughput / (2 ** 20)), int(iops)) for num_threads, throughput, iops in zip(num_threads_list, throughputs, IOPSs)))

Benchmark results with 4K byte feature dimension:

(venv) mfbalin@BALIN-PC:~/dgl-1$ python graphbolt/benchmarks/disk_based_feature.py
1 715 MiB/s 183271 IOPS
2 1142 MiB/s 292576 IOPS
3 1571 MiB/s 402287 IOPS
4 1785 MiB/s 457213 IOPS
5 1873 MiB/s 479614 IOPS
6 2094 MiB/s 536154 IOPS
7 2185 MiB/s 559578 IOPS
8 2228 MiB/s 570410 IOPS
[(1, 715, 183271), (2, 1142, 292576), (3, 1571, 402287), (4, 1785, 457213), (5, 1873, 479614), (6, 2094, 536154), (7, 2185, 559578), (8, 2228, 570410)]

Checklist

Please feel free to remove inapplicable items for your PR.

• The PR title starts with [$CATEGORY] (such as [NN], [Model], [Doc], [Feature]])
• I've leverage the tools to beautify the python and c++ code.
• The PR is complete and small, read the Google eng practice (CL equals to PR) to understand more about small PR. In DGL, we consider PRs with less than 200 lines of core code change are small (example, test and documentation could be exempted).
• All changes have test coverage
• Code is well-documented
• To the best of my knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change
• Related issue is referred in this PR
• If the PR is for a new model/paper, I've updated the example index here.

Changes

[dgl-bot]

To trigger regression tests:

• @dgl-bot run [instance-type] [which tests] [compare-with-branch];
  For example: @dgl-bot run g4dn.4xlarge all dmlc/master or @dgl-bot run c5.9xlarge kernel,api dmlc/master

[dgl-bot]

Commit ID: abba7f6

Build ID: 1

Status: ⚪️ CI test cancelled due to overrun.

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: bfe1981

Build ID: 2

Status: ✅ CI test succeeded.

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: 45d3522

Build ID: 3

Status: ❌ CI test failed in Stage [CPU Build (Win64)].

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: 0fb9a5b

Build ID: 4

Status: ⚪️ CI test cancelled due to overrun.

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: 82f1013

Build ID: 5

Status: ⚪️ CI test cancelled due to overrun.

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: a60d979

Build ID: 6

Status: ⚪️ CI test cancelled due to overrun.

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: 2a858db

Build ID: 7

Status: ❌ CI test failed in Stage [CPU Build].

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: 01b0b84

Build ID: 8

Status: ❌ CI test failed in Stage [CPU Build (Win64)].

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: 8d9c8df

Build ID: 9

Status: ❌ CI test failed in Stage [CPU Build (Win64)].

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: e8ee557

Build ID: 10

Status: ✅ CI test succeeded.

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: ee36568

Build ID: 11

Status: ✅ CI test succeeded.

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: ca8da895f52b48eb340b65da409e8b7e150785a5

Build ID: 12

Status: ✅ CI test succeeded.

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: 4c877c6

Build ID: 24

Status: ❌ CI test failed in Stage [CPU Build].

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: e9e9dd7

Build ID: 25

Status: ✅ CI test succeeded.

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: 1affb480c5118bf4190a05a3639122f13000f15a

Build ID: 26

Status: ❌ CI test failed in Stage [Torch CPU Example test].

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: e4f93ae

Build ID: 27

Status: ⚪️ CI test cancelled due to overrun.

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: c84a7f0a13c51d43f52a01ebf50e7d6b64cb4aec

Build ID: 28

Status: ⚪️ CI test cancelled due to overrun.

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: fcb09ed36b6800cc994ab6556dde2985463206e1

Build ID: 29

Status: ⚪️ CI test cancelled due to overrun.

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: 4167fa339f3986b72ed871f033467248b035e09a

Build ID: 30

Status: ⚪️ CI test cancelled due to overrun.

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: 68b75661a829a875e2da6f93f3d01e6283b055ca

Build ID: 31

Status: ⚪️ CI test cancelled due to overrun.

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: 042df7debdfbe00d61f838f6b4ab172e551222dc

Build ID: 32

Status: ❌ CI test failed in Stage [CPU Build].

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: e596c721a01d6dabdcdfffd3f50f4627d6709e78

Build ID: 33

Status: ❌ CI test failed in Stage [CPU Build].

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: 6c5c035

Build ID: 34

Status: ✅ CI test succeeded.

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: ba7c89bda3077c7da1e07767885d4e1f3784317e

Build ID: 35

Status: ✅ CI test succeeded.

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: 0b2a630

Build ID: 36

Status: ✅ CI test succeeded.

Report path: link

Full logs path: link

[dgl-bot]

Commit ID: ebf0e2b

Build ID: 37

Status: ✅ CI test succeeded.

Report path: link

Full logs path: link