long nbLines = files.count.get() * LINES_PER_FILE;
long throughputPerSecond = 1000L * nbLines / period;
LOGGER.info(String.format("%d lines indexed in %d ms (%d docs/second)", nbLines, period, throughputPerSecond));
- benchmark.expectBetween("Throughput to index source lines", throughputPerSecond, 7800, 8200);
+ benchmark.expectBetween("Throughput to index source lines", throughputPerSecond, 7500, 8000);
// be sure that physical files do not evolve during estimation of size
tester.get(EsClient.class).prepareOptimize(SourceLineIndexDefinition.INDEX).get();