• There are 914 files with 97,379 lines of code.
• 11 very long files (15,928 lines of code)
• 20 long files (14,953 lines of code)
• 85 medium size files (24,381 lines of code)
• 133 small files (18,547 lines of code)
• 665 very small files (23,570 lines of code)

File	# lines	≈ tokens to read	# changes (1y)	≈ lines edited per change	lines read per line changed
engine.rs in src/storage/chunk_engine/src/core	1451	~10k–30k	3	~15	~100
IBSocket.cc in src/common/net/ib	1073	~7k–20k	1	~15	~70
cxx.rs in src/storage/chunk_engine/src	536	~3k–10k	2	~7	~70
StorageOperator.cc in src/storage/service	1016	~7k–15k	1	~30	~30
Common.h in src/fbs/storage	665	~3k–10k	1	~5	~100
IBConnect.cc in src/common/net/ib	646	~3k–10k	1	~10	~50
IBSocket.h in src/common/net/ib	467	~3k–7k	1	~2	~300
StorageClient.h in src/client/storage	456	~3k–7k	1	~2	~200
hf3fs_fuse_main.toml in configs	393	~2k–7k	1	~2	~200
rocksdb.rs in src/storage/chunk_engine/src/meta	260	~1.5k–5k	1	~2	~150
ReliableForwarding.cc in src/storage/service	242	~1.5k–5k	1	~5	~50
IoRing.cc in src/fuse	240	~1.5k–5k	1	~2	~100
IOWorker.cc in src/common/net	112	~700–2k	2	~10	~10
AtomicSharedPtrTable.h in src/common/utils	65	~300–1k	3	~30	~2
TransportPool.cc in src/common/net	186	~1k–3k	1	~3	~50
Schema.cc in src/fbs/meta	180	~1k–3k	1	~10	~20
allocators.rs in src/storage/chunk_engine/src/alloc	178	~1k–3k	1	~2	~100
Shuffle.h in src/common/utils	157	~1k–3k	1	~200	~1
cli.py in hf3fs_utils	141	~700–2k	1	-	-
SetChainsOperation.cc in src/mgmtd/ops	124	~700–2k	1	~2	~70

Tokens are lines × 7 to 14 (analysis.fileReadsForChanges). Lines edited per change are the file's lines added and deleted divided by its commits over the whole history. Agents read about 2,000 lines at a time: when a 17,000-line file was split into modules, the tokens an agent read for the same change fell by 83% (The Economic Benefit of Refactoring, 2026).

There are 8 files with lines longer than 120 characters. In total, there are 72 long lines.

File	# lines	# units	# long lines
data_placement.py in deploy/data_placement/src/model	449	35	39
error_definitions.h in src/fdb	259	-	11
data_placement_job.py in deploy/data_placement/src/model	89	4	9
gen_chain_table.py in deploy/data_placement/src/setup	100	3	6
hf3fs_usrbio.cpp in benchmarks/fio_usrbio	247	12	3
in VersionInfo.cc.in in src/common/utils	42	-	2
Recorder.h in src/common/monitor	345	30	1
IBSocket.cc in src/common/net/ib	1073	52	1

• File size measurements show the distribution of size of files.
• Files are classified in four categories based on their size (lines of code): 1-100 (very small files), 101-200 (small files), 201-500 (medium size files), 501-1000 (long files), 1001+(very long files).
• It is a good practice to keep files small. Long files may become "bloaters", code that have increased to such gargantuan proportions that they are hard to work with.