diff --git a/scripts/comparator/run.sh b/scripts/comparator/run.sh
index d1bb833188..ea5c05d781 100755
--- a/scripts/comparator/run.sh
+++ b/scripts/comparator/run.sh
@@ -75,4 +75,4 @@ $comparator -input=$QUERY_FILE \
 -regressionDir=$REGRESSION_DIR
 
 # Run PromQL testdata tests
-go test -v -timeout 300s -tags=compatibility -count=1 github.com/m3db/m3/src/query/test/
+go test -v -timeout 300s -tags=compatibility -count=1 github.com/m3db/m3/src/query/test/compatibility/
diff --git a/src/cmd/services/m3comparator/main/querier.go b/src/cmd/services/m3comparator/main/querier.go
index 41229b983d..91cab6eb50 100644
--- a/src/cmd/services/m3comparator/main/querier.go
+++ b/src/cmd/services/m3comparator/main/querier.go
@@ -36,6 +36,7 @@ import (
 	"github.com/m3db/m3/src/dbnode/encoding"
 	"github.com/m3db/m3/src/dbnode/ts"
 	"github.com/m3db/m3/src/query/block"
+	"github.com/m3db/m3/src/query/models"
 	"github.com/m3db/m3/src/query/storage"
 	"github.com/m3db/m3/src/query/storage/m3"
 	"github.com/m3db/m3/src/query/storage/m3/consolidators"
@@ -149,27 +150,38 @@ func (q *querier) FetchCompressed(
 	options *storage.FetchOptions,
 ) (consolidators.SeriesFetchResult, m3.Cleanup, error) {
 	var (
-		iters        encoding.SeriesIterators
-		randomSeries []series
-		ignoreFilter bool
-		err          error
-		nameTagFound bool
+		iters               encoding.SeriesIterators
+		randomSeries        []series
+		ignoreFilter        bool
+		err                 error
+		strictMetricsFilter bool
 	)
 
 	name := q.iteratorOpts.tagOptions.MetricName()
 	for _, matcher := range query.TagMatchers {
 		if bytes.Equal(name, matcher.Name) {
-			nameTagFound = true
-			iters, err = q.handler.getSeriesIterators(string(matcher.Value))
-			if err != nil {
-				return consolidators.SeriesFetchResult{}, noop, err
+
+			metricsName := string(matcher.Value)
+
+			// NB: the default behaviour of this querier is to return predefined metrics with random data if no match by
+			// metrics name is found. To force it return an empty result, query the "nonexistent*" metrics.
+			if match, _ := regexp.MatchString("^nonexist[ae]nt", metricsName); match {
+				return consolidators.SeriesFetchResult{}, noop, nil
 			}
 
-			break
+			if matcher.Type == models.MatchEqual {
+				strictMetricsFilter = true
+				iters, err = q.handler.getSeriesIterators(metricsName)
+				if err != nil {
+					return consolidators.SeriesFetchResult{}, noop, err
+				}
+
+				break
+			}
 		}
 	}
 
-	if iters == nil && !nameTagFound && len(query.TagMatchers) > 0 {
+	if iters == nil && !strictMetricsFilter && len(query.TagMatchers) > 0 {
 		iters, err = q.handler.getSeriesIterators("")
 		if err != nil {
 			return consolidators.SeriesFetchResult{}, noop, err
diff --git a/src/cmd/services/m3comparator/main/querier_test.go b/src/cmd/services/m3comparator/main/querier_test.go
index 574f9354e5..69d73e4438 100644
--- a/src/cmd/services/m3comparator/main/querier_test.go
+++ b/src/cmd/services/m3comparator/main/querier_test.go
@@ -111,6 +111,16 @@ func TestGenerateRandomSeries(t *testing.T) {
 		givenQuery *storage.FetchQuery
 		wantSeries []tagMap
 	}{
+		{
+			name:       "querying nonexistent_metric returns empty",
+			givenQuery: matcherQuery(t, metricNameTag, "nonexistent_metric"),
+			wantSeries: []tagMap{},
+		},
+		{
+			name:       "querying nonexistant returns empty",
+			givenQuery: matcherQuery(t, metricNameTag, "nonexistant"),
+			wantSeries: []tagMap{},
+		},
 		{
 			name:       "random data for known metrics",
 			givenQuery: matcherQuery(t, metricNameTag, "quail"),
diff --git a/src/cmd/services/m3comparator/main/series_load_handler.go b/src/cmd/services/m3comparator/main/series_load_handler.go
index 8c60f8c495..cdfdbe0e04 100644
--- a/src/cmd/services/m3comparator/main/series_load_handler.go
+++ b/src/cmd/services/m3comparator/main/series_load_handler.go
@@ -88,64 +88,75 @@ func (l *httpSeriesLoadHandler) getSeriesIterators(
 	l.RLock()
 	defer l.RUnlock()
 
+	var seriesMaps []idSeriesMap
 	logger := l.iterOpts.iOpts.Logger()
-	seriesMap, found := l.nameIDSeriesMap[name]
-	if !found || len(seriesMap.series) == 0 {
-		return nil, nil
+	if name == "" {
+		// return all preloaded data
+		for _, series := range l.nameIDSeriesMap {
+			seriesMaps = append(seriesMaps, series)
+		}
+	} else {
+		seriesMap, found := l.nameIDSeriesMap[name]
+		if !found || len(seriesMap.series) == 0 {
+			return nil, nil
+		}
+		seriesMaps = append(seriesMaps, seriesMap)
 	}
 
-	iters := make([]encoding.SeriesIterator, 0, len(seriesMap.series))
-	for _, series := range seriesMap.series {
-		encoder := l.iterOpts.encoderPool.Get()
-		dps := series.Datapoints
-		startTime := time.Time{}
-		if len(dps) > 0 {
-			startTime = dps[0].Timestamp.Truncate(time.Hour)
-		}
+	iters := make([]encoding.SeriesIterator, 0, len(seriesMaps))
+	for _, seriesMap := range seriesMaps {
+		for _, series := range seriesMap.series {
+			encoder := l.iterOpts.encoderPool.Get()
+			dps := series.Datapoints
+			startTime := time.Time{}
+			if len(dps) > 0 {
+				startTime = dps[0].Timestamp.Truncate(time.Hour)
+			}
 
-		encoder.Reset(startTime, len(dps), nil)
-		for _, dp := range dps {
-			err := encoder.Encode(ts.Datapoint{
-				Timestamp:      dp.Timestamp,
-				Value:          float64(dp.Value),
-				TimestampNanos: xtime.ToUnixNano(dp.Timestamp),
-			}, xtime.Nanosecond, nil)
-
-			if err != nil {
-				encoder.Close()
-				logger.Error("error encoding datapoints", zap.Error(err))
-				return nil, err
+			encoder.Reset(startTime, len(dps), nil)
+			for _, dp := range dps {
+				err := encoder.Encode(ts.Datapoint{
+					Timestamp:      dp.Timestamp,
+					Value:          float64(dp.Value),
+					TimestampNanos: xtime.ToUnixNano(dp.Timestamp),
+				}, xtime.Nanosecond, nil)
+
+				if err != nil {
+					encoder.Close()
+					logger.Error("error encoding datapoints", zap.Error(err))
+					return nil, err
+				}
 			}
-		}
 
-		readers := [][]xio.BlockReader{{{
-			SegmentReader: xio.NewSegmentReader(encoder.Discard()),
-			Start:         series.Start,
-			BlockSize:     series.End.Sub(series.Start),
-		}}}
-
-		multiReader := encoding.NewMultiReaderIterator(
-			iterAlloc,
-			l.iterOpts.iteratorPools.MultiReaderIterator(),
-		)
-
-		sliceOfSlicesIter := xio.NewReaderSliceOfSlicesFromBlockReadersIterator(readers)
-		multiReader.ResetSliceOfSlices(sliceOfSlicesIter, nil)
-
-		tagIter, id := buildTagIteratorAndID(series.Tags, l.iterOpts.tagOptions)
-		iter := encoding.NewSeriesIterator(
-			encoding.SeriesIteratorOptions{
-				ID:             id,
-				Namespace:      ident.StringID("ns"),
-				Tags:           tagIter,
-				StartInclusive: xtime.ToUnixNano(series.Start),
-				EndExclusive:   xtime.ToUnixNano(series.End),
-				Replicas: []encoding.MultiReaderIterator{
-					multiReader,
-				},
-			}, nil)
-
-		iters = append(iters, iter)
+			readers := [][]xio.BlockReader{{{
+				SegmentReader: xio.NewSegmentReader(encoder.Discard()),
+				Start:         series.Start,
+				BlockSize:     series.End.Sub(series.Start),
+			}}}
+
+			multiReader := encoding.NewMultiReaderIterator(
+				iterAlloc,
+				l.iterOpts.iteratorPools.MultiReaderIterator(),
+			)
+
+			sliceOfSlicesIter := xio.NewReaderSliceOfSlicesFromBlockReadersIterator(readers)
+			multiReader.ResetSliceOfSlices(sliceOfSlicesIter, nil)
+
+			tagIter, id := buildTagIteratorAndID(series.Tags, l.iterOpts.tagOptions)
+			iter := encoding.NewSeriesIterator(
+				encoding.SeriesIteratorOptions{
+					ID:             id,
+					Namespace:      ident.StringID("ns"),
+					Tags:           tagIter,
+					StartInclusive: xtime.ToUnixNano(series.Start),
+					EndExclusive:   xtime.ToUnixNano(series.End),
+					Replicas: []encoding.MultiReaderIterator{
+						multiReader,
+					},
+				}, nil)
+
+			iters = append(iters, iter)
+		}
 	}
 
 	return encoding.NewSeriesIterators(
diff --git a/src/cmd/services/m3comparator/main/series_load_handler_test.go b/src/cmd/services/m3comparator/main/series_load_handler_test.go
index ec01a83bcd..87df74ec5c 100644
--- a/src/cmd/services/m3comparator/main/series_load_handler_test.go
+++ b/src/cmd/services/m3comparator/main/series_load_handler_test.go
@@ -146,23 +146,33 @@ const seriesStr = `
     }
 ]`
 
-func TestIngestSeries(t *testing.T) {
-	opts := iteratorOptions{
-		encoderPool:   encoderPool,
-		iteratorPools: iterPools,
-		tagOptions:    tagOptions,
-		iOpts:         iOpts,
+const otherSeriesStr = `
+[
+    {
+        "start": "2020-03-30T12:00:00Z",
+        "end": "2020-03-30T12:01:00Z",
+        "tags": [
+            ["__name__", "foo"],
+            ["bar", "baz"]
+        ],
+        "datapoints": [
+            { "val": "555", "ts": "2020-03-30T12:00:00Z" }
+		]
 	}
+]`
 
-	req, err := http.NewRequest(http.MethodPost, "", strings.NewReader(seriesStr))
-	require.NoError(t, err)
-
-	recorder := httptest.NewRecorder()
+var opts = iteratorOptions{
+	encoderPool:   encoderPool,
+	iteratorPools: iterPools,
+	tagOptions:    tagOptions,
+	iOpts:         iOpts,
+}
 
+func TestIngestSeries(t *testing.T) {
 	handler := newHTTPSeriesLoadHandler(opts)
-	handler.ServeHTTP(recorder, req)
 
-	assert.Equal(t, http.StatusOK, recorder.Code)
+	loadSeries(t, handler, seriesStr)
+	loadSeries(t, handler, otherSeriesStr)
 
 	iters, err := handler.getSeriesIterators("series_name")
 	require.NoError(t, err)
@@ -173,7 +183,7 @@ func TestIngestSeries(t *testing.T) {
 	require.Equal(t, 1, len(expectedList))
 	expected := expectedList[0]
 
-	require.Equal(t, 1, len(iters.Iters()))
+	require.Equal(t, 1, iters.Len())
 	it := iters.Iters()[0]
 	j := 0
 	for it.Next() {
@@ -190,38 +200,38 @@ func TestIngestSeries(t *testing.T) {
 	assert.Equal(t, j, len(expected.Datapoints))
 }
 
-func TestClearData(t *testing.T) {
-	opts := iteratorOptions{
-		encoderPool:   encoderPool,
-		iteratorPools: iterPools,
-		tagOptions:    tagOptions,
-		iOpts:         iOpts,
-	}
+func TestGetAllSeries(t *testing.T) {
+	handler := newHTTPSeriesLoadHandler(opts)
 
-	req, err := http.NewRequest(http.MethodPost, "", strings.NewReader(seriesStr))
-	require.NoError(t, err)
+	loadSeries(t, handler, seriesStr)
+	loadSeries(t, handler, otherSeriesStr)
 
-	recorder := httptest.NewRecorder()
+	iters, err := handler.getSeriesIterators("")
+	require.NoError(t, err)
+	require.Equal(t, 2, iters.Len())
+}
 
+func TestClearData(t *testing.T) {
 	handler := newHTTPSeriesLoadHandler(opts)
-	handler.ServeHTTP(recorder, req)
 
-	assert.Equal(t, http.StatusOK, recorder.Code)
+	loadSeries(t, handler, seriesStr)
+	loadSeries(t, handler, otherSeriesStr)
 
-	iters, err := handler.getSeriesIterators("series_name")
+	iters, err := handler.getSeriesIterators("")
 	require.NoError(t, err)
-	require.Equal(t, 1, len(iters.Iters()))
+	require.Equal(t, 2, iters.Len())
 
 	// Call clear data
-	req, err = http.NewRequest(http.MethodDelete, "", nil)
+	req, err := http.NewRequest(http.MethodDelete, "", nil)
 	require.NoError(t, err)
 
+	recorder := httptest.NewRecorder()
 	handler.ServeHTTP(recorder, req)
 	assert.Equal(t, http.StatusOK, recorder.Code)
 
-	iters, err = handler.getSeriesIterators("series_name")
+	iters, err = handler.getSeriesIterators("")
 	require.NoError(t, err)
-	require.Nil(t, iters)
+	require.Equal(t, 0, iters.Len())
 }
 
 func readTags(it encoding.SeriesIterator) parser.Tags {
@@ -235,3 +245,14 @@ func readTags(it encoding.SeriesIterator) parser.Tags {
 
 	return tags
 }
+
+func loadSeries(t *testing.T, handler *httpSeriesLoadHandler, data string) {
+	req, err := http.NewRequest(http.MethodPost, "", strings.NewReader(data))
+	require.NoError(t, err)
+
+	recorder := httptest.NewRecorder()
+
+	handler.ServeHTTP(recorder, req)
+
+	assert.Equal(t, http.StatusOK, recorder.Code)
+}
diff --git a/src/query/api/v1/handler/prometheus/response.go b/src/query/api/v1/handler/prometheus/response.go
index d4cbb9bed2..a28384e162 100644
--- a/src/query/api/v1/handler/prometheus/response.go
+++ b/src/query/api/v1/handler/prometheus/response.go
@@ -48,23 +48,23 @@ type result interface {
 	matches(other result) (MatchInformation, error)
 }
 
-type matrixResult struct {
-	// Result is the list of matrixRow for the response.
+// MatrixResult contains a list matrixRow.
+type MatrixResult struct {
 	Result []matrixRow `json:"result"`
 }
 
-type vectorResult struct {
-	// Result is the list of vectorItem for the response.
+// VectorResult contains a list of vectorItem.
+type VectorResult struct {
 	Result []vectorItem `json:"result"`
 }
 
-type scalarResult struct {
-	// Result is the scalar Value for the response.
+// ScalarResult is the scalar Value for the response.
+type ScalarResult struct {
 	Result Value `json:"result"`
 }
 
-type stringResult struct {
-	// Result is the string Value for the response.
+// StringResult is the string Value for the response.
+type StringResult struct {
 	Result Value `json:"result"`
 }
 
@@ -81,16 +81,16 @@ func (d *data) UnmarshalJSON(bytes []byte) error {
 	switch discriminator.ResultType {
 
 	case "matrix":
-		d.Result = &matrixResult{}
+		d.Result = &MatrixResult{}
 
 	case "vector":
-		d.Result = &vectorResult{}
+		d.Result = &VectorResult{}
 
 	case "scalar":
-		d.Result = &scalarResult{}
+		d.Result = &ScalarResult{}
 
 	case "string":
-		d.Result = &stringResult{}
+		d.Result = &StringResult{}
 
 	default:
 		return fmt.Errorf("unknown resultType: %s", discriminator.ResultType)
@@ -100,19 +100,19 @@ func (d *data) UnmarshalJSON(bytes []byte) error {
 }
 
 // Len is the number of elements in the collection.
-func (r matrixResult) Len() int { return len(r.Result) }
+func (r MatrixResult) Len() int { return len(r.Result) }
 
 // Less reports whether the element with
 // index i should sort before the element with index j.
-func (r matrixResult) Less(i, j int) bool {
+func (r MatrixResult) Less(i, j int) bool {
 	return r.Result[i].id < r.Result[j].id
 }
 
 // Swap swaps the elements with indexes i and j.
-func (r matrixResult) Swap(i, j int) { r.Result[i], r.Result[j] = r.Result[j], r.Result[i] }
+func (r MatrixResult) Swap(i, j int) { r.Result[i], r.Result[j] = r.Result[j], r.Result[i] }
 
-// Sort sorts the matrixResult.
-func (r matrixResult) Sort() {
+// Sort sorts the MatrixResult.
+func (r MatrixResult) Sort() {
 	for i, result := range r.Result {
 		r.Result[i].id = result.Metric.genID()
 	}
@@ -121,19 +121,19 @@ func (r matrixResult) Sort() {
 }
 
 // Len is the number of elements in the vector.
-func (r vectorResult) Len() int { return len(r.Result) }
+func (r VectorResult) Len() int { return len(r.Result) }
 
 // Less reports whether the element with
 // index i should sort before the element with index j.
-func (r vectorResult) Less(i, j int) bool {
+func (r VectorResult) Less(i, j int) bool {
 	return r.Result[i].id < r.Result[j].id
 }
 
 // Swap swaps the elements with indexes i and j.
-func (r vectorResult) Swap(i, j int) { r.Result[i], r.Result[j] = r.Result[j], r.Result[i] }
+func (r VectorResult) Swap(i, j int) { r.Result[i], r.Result[j] = r.Result[j], r.Result[i] }
 
-// Sort sorts the vectorResult.
-func (r vectorResult) Sort() {
+// Sort sorts the VectorResult.
+func (r VectorResult) Sort() {
 	for i, result := range r.Result {
 		r.Result[i].id = result.Metric.genID()
 	}
@@ -224,10 +224,10 @@ func (d data) matches(other data) (MatchInformation, error) {
 	return d.Result.matches(other.Result)
 }
 
-func (r matrixResult) matches(other result) (MatchInformation, error) {
-	otherMatrix, ok := other.(*matrixResult)
+func (r MatrixResult) matches(other result) (MatchInformation, error) {
+	otherMatrix, ok := other.(*MatrixResult)
 	if !ok {
-		err := fmt.Errorf("incorrect type for matching, expected matrixResult, %v", other)
+		err := fmt.Errorf("incorrect type for matching, expected MatrixResult, %v", other)
 		return MatchInformation{
 			NoMatch: true,
 		}, err
@@ -254,10 +254,10 @@ func (r matrixResult) matches(other result) (MatchInformation, error) {
 	return MatchInformation{FullMatch: true}, nil
 }
 
-func (r vectorResult) matches(other result) (MatchInformation, error) {
-	otherVector, ok := other.(*vectorResult)
+func (r VectorResult) matches(other result) (MatchInformation, error) {
+	otherVector, ok := other.(*VectorResult)
 	if !ok {
-		err := fmt.Errorf("incorrect type for matching, expected vectorResult")
+		err := fmt.Errorf("incorrect type for matching, expected VectorResult")
 		return MatchInformation{
 			NoMatch: true,
 		}, err
@@ -284,10 +284,10 @@ func (r vectorResult) matches(other result) (MatchInformation, error) {
 	return MatchInformation{FullMatch: true}, nil
 }
 
-func (r scalarResult) matches(other result) (MatchInformation, error) {
-	otherScalar, ok := other.(*scalarResult)
+func (r ScalarResult) matches(other result) (MatchInformation, error) {
+	otherScalar, ok := other.(*ScalarResult)
 	if !ok {
-		err := fmt.Errorf("incorrect type for matching, expected scalarResult")
+		err := fmt.Errorf("incorrect type for matching, expected ScalarResult")
 		return MatchInformation{
 			NoMatch: true,
 		}, err
@@ -302,10 +302,10 @@ func (r scalarResult) matches(other result) (MatchInformation, error) {
 	return MatchInformation{FullMatch: true}, nil
 }
 
-func (r stringResult) matches(other result) (MatchInformation, error) {
-	otherString, ok := other.(*stringResult)
+func (r StringResult) matches(other result) (MatchInformation, error) {
+	otherString, ok := other.(*StringResult)
 	if !ok {
-		err := fmt.Errorf("incorrect type for matching, expected stringResult")
+		err := fmt.Errorf("incorrect type for matching, expected StringResult")
 		return MatchInformation{
 			NoMatch: true,
 		}, err
diff --git a/src/query/api/v1/handler/prometheus/response_test.go b/src/query/api/v1/handler/prometheus/response_test.go
index a067822ffd..c076410458 100644
--- a/src/query/api/v1/handler/prometheus/response_test.go
+++ b/src/query/api/v1/handler/prometheus/response_test.go
@@ -64,7 +64,7 @@ func TestUnmarshalPrometheusResponse(t *testing.T) {
 				"success",
 				data{
 					"scalar",
-					&scalarResult{Value{1590605774.0, "84"}},
+					&ScalarResult{Value{1590605774.0, "84"}},
 				},
 			},
 		},
@@ -81,7 +81,7 @@ func TestUnmarshalPrometheusResponse(t *testing.T) {
 				"success",
 				data{
 					"string",
-					&stringResult{Value{1590605775.0, "FOO"}},
+					&StringResult{Value{1590605775.0, "FOO"}},
 				},
 			},
 		},
@@ -113,7 +113,7 @@ func TestUnmarshalPrometheusResponse(t *testing.T) {
 				"success",
 				data{
 					"vector",
-					&vectorResult{[]vectorItem{
+					&VectorResult{[]vectorItem{
 						{
 							Metric: Tags{"__name__": "foo", "bar": "1"},
 							Value:  Value{1590605775.0, "0.5"},
@@ -154,7 +154,7 @@ func TestUnmarshalPrometheusResponse(t *testing.T) {
 				"success",
 				data{
 					"matrix",
-					&matrixResult{[]matrixRow{
+					&MatrixResult{[]matrixRow{
 						{
 							Metric: Tags{"__name__": "foo", "bar": "1"},
 							Values: Values{{1590605775.0, "1"}, {1590605785.0, "11"}},
@@ -197,7 +197,7 @@ func TestResponseMatching(t *testing.T) {
 				"success",
 				data{
 					"scalar",
-					&scalarResult{Value{1590605774.0, "1"}},
+					&ScalarResult{Value{1590605774.0, "1"}},
 				},
 			},
 		},
@@ -207,7 +207,7 @@ func TestResponseMatching(t *testing.T) {
 				"success",
 				data{
 					"scalar",
-					&scalarResult{Value{1590605775.0, "1"}},
+					&ScalarResult{Value{1590605775.0, "1"}},
 				},
 			},
 		},
@@ -217,7 +217,7 @@ func TestResponseMatching(t *testing.T) {
 				"success",
 				data{
 					"scalar",
-					&scalarResult{Value{1590605774.0, "2"}},
+					&ScalarResult{Value{1590605774.0, "2"}},
 				},
 			},
 		},
@@ -228,7 +228,7 @@ func TestResponseMatching(t *testing.T) {
 				"success",
 				data{
 					"vector",
-					&vectorResult{[]vectorItem{
+					&VectorResult{[]vectorItem{
 						{
 							Metric: Tags{"__name__": "foo"},
 							Value:  Value{1590605775.0, "0.5"},
@@ -243,7 +243,7 @@ func TestResponseMatching(t *testing.T) {
 				"success",
 				data{
 					"vector",
-					&vectorResult{[]vectorItem{
+					&VectorResult{[]vectorItem{
 						{
 							Metric: Tags{"__name__": "foo", "bar": "1"},
 							Value:  Value{1590605775.0, "0.5"},
@@ -258,7 +258,7 @@ func TestResponseMatching(t *testing.T) {
 				"success",
 				data{
 					"vector",
-					&vectorResult{[]vectorItem{
+					&VectorResult{[]vectorItem{
 						{
 							Metric: Tags{"__name__": "foo", "bar": "1"},
 							Value:  Value{1590605775.0, "0.5"},
@@ -277,7 +277,7 @@ func TestResponseMatching(t *testing.T) {
 				"success",
 				data{
 					"vector",
-					&vectorResult{[]vectorItem{
+					&VectorResult{[]vectorItem{
 						{
 							Metric: Tags{"__name__": "bar"},
 							Value:  Value{1590605775.0, "0.5"},
@@ -292,7 +292,7 @@ func TestResponseMatching(t *testing.T) {
 				"success",
 				data{
 					"vector",
-					&vectorResult{[]vectorItem{
+					&VectorResult{[]vectorItem{
 						{
 							Metric: Tags{"__name__": "foo"},
 							Value:  Value{1590605775.0, "1"},
@@ -307,7 +307,7 @@ func TestResponseMatching(t *testing.T) {
 				"success",
 				data{
 					"vector",
-					&vectorResult{[]vectorItem{
+					&VectorResult{[]vectorItem{
 						{
 							Metric: Tags{"__name__": "foo"},
 							Value:  Value{1590605774.0, "0.5"},
@@ -323,7 +323,7 @@ func TestResponseMatching(t *testing.T) {
 				"success",
 				data{
 					"matrix",
-					&matrixResult{[]matrixRow{
+					&MatrixResult{[]matrixRow{
 						{
 							Metric: Tags{"__name__": "foo"},
 							Values: Values{{1590605775.0, "1"}},
@@ -338,7 +338,7 @@ func TestResponseMatching(t *testing.T) {
 				"success",
 				data{
 					"matrix",
-					&matrixResult{[]matrixRow{
+					&MatrixResult{[]matrixRow{
 						{
 							Metric: Tags{"__name__": "bar"},
 							Values: Values{{1590605775.0, "1"}},
@@ -353,7 +353,7 @@ func TestResponseMatching(t *testing.T) {
 				"success",
 				data{
 					"matrix",
-					&matrixResult{[]matrixRow{
+					&MatrixResult{[]matrixRow{
 						{
 							Metric: Tags{"__name__": "foo"},
 							Values: Values{{1590605775.0, "2"}},
@@ -368,7 +368,7 @@ func TestResponseMatching(t *testing.T) {
 				"success",
 				data{
 					"matrix",
-					&matrixResult{[]matrixRow{
+					&MatrixResult{[]matrixRow{
 						{
 							Metric: Tags{"__name__": "foo"},
 							Values: Values{{1590605776.0, "1"}},
@@ -383,7 +383,7 @@ func TestResponseMatching(t *testing.T) {
 				"success",
 				data{
 					"matrix",
-					&matrixResult{[]matrixRow{
+					&MatrixResult{[]matrixRow{
 						{
 							Metric: Tags{"__name__": "foo", "bar": "1"},
 							Values: Values{{1590605775.0, "1"}},
@@ -398,7 +398,7 @@ func TestResponseMatching(t *testing.T) {
 				"success",
 				data{
 					"matrix",
-					&matrixResult{[]matrixRow{
+					&MatrixResult{[]matrixRow{
 						{
 							Metric: Tags{"__name__": "foo"},
 							Values: Values{{1590605775.0, "1"}, {1590605776.0, "2"}},
@@ -413,7 +413,7 @@ func TestResponseMatching(t *testing.T) {
 				"success",
 				data{
 					"matrix",
-					&matrixResult{[]matrixRow{
+					&MatrixResult{[]matrixRow{
 						{
 							Metric: Tags{"__name__": "foo"},
 							Values: Values{{1590605775.0, "1"}},
@@ -456,7 +456,7 @@ func TestResponseMatchingOrderInsensitive(t *testing.T) {
 				"success",
 				data{
 					"vector",
-					&vectorResult{[]vectorItem{
+					&VectorResult{[]vectorItem{
 						{
 							Metric: Tags{"__name__": "first"},
 							Value:  Value{1590605775.0, "1"},
@@ -472,7 +472,7 @@ func TestResponseMatchingOrderInsensitive(t *testing.T) {
 				"success",
 				data{
 					"vector",
-					&vectorResult{[]vectorItem{
+					&VectorResult{[]vectorItem{
 						{
 							Metric: Tags{"__name__": "second"},
 							Value:  Value{1590605775.0, "2"},
@@ -491,7 +491,7 @@ func TestResponseMatchingOrderInsensitive(t *testing.T) {
 				"success",
 				data{
 					"matrix",
-					&matrixResult{[]matrixRow{
+					&MatrixResult{[]matrixRow{
 						{
 							Metric: Tags{"__name__": "first"},
 							Values: Values{{1590605775.0, "1"}},
@@ -507,7 +507,7 @@ func TestResponseMatchingOrderInsensitive(t *testing.T) {
 				"success",
 				data{
 					"matrix",
-					&matrixResult{[]matrixRow{
+					&MatrixResult{[]matrixRow{
 						{
 							Metric: Tags{"__name__": "second"},
 							Values: Values{{1590605775.0, "2"}},
diff --git a/src/query/test/m3comparator_client.go b/src/query/test/compatibility/m3comparator_client.go
similarity index 99%
rename from src/query/test/m3comparator_client.go
rename to src/query/test/compatibility/m3comparator_client.go
index a3582909fb..58819d853b 100644
--- a/src/query/test/m3comparator_client.go
+++ b/src/query/test/compatibility/m3comparator_client.go
@@ -18,7 +18,7 @@
 // OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 // THE SOFTWARE.
 
-package test
+package compatibility
 
 import (
 	"bytes"
diff --git a/src/query/test/m3query_client.go b/src/query/test/compatibility/m3query_client.go
similarity index 98%
rename from src/query/test/m3query_client.go
rename to src/query/test/compatibility/m3query_client.go
index 8b0c57bad1..067f8faa57 100644
--- a/src/query/test/m3query_client.go
+++ b/src/query/test/compatibility/m3query_client.go
@@ -18,7 +18,7 @@
 // OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 // THE SOFTWARE.
 
-package test
+package compatibility
 
 import (
 	"fmt"
diff --git a/src/query/test/promql_test.go b/src/query/test/compatibility/promql_test.go
similarity index 99%
rename from src/query/test/promql_test.go
rename to src/query/test/compatibility/promql_test.go
index 7dab41d803..d5f756800f 100644
--- a/src/query/test/promql_test.go
+++ b/src/query/test/compatibility/promql_test.go
@@ -35,7 +35,7 @@
 
 // This code was taken from prometheus repo: https://github.com/prometheus/prometheus/blob/master/promql/promql_test.go
 
-package test
+package compatibility
 
 import (
 	"path/filepath"
diff --git a/src/query/test/test.go b/src/query/test/compatibility/test.go
similarity index 91%
rename from src/query/test/test.go
rename to src/query/test/compatibility/test.go
index 97dd3b52a3..f67600c8f3 100644
--- a/src/query/test/test.go
+++ b/src/query/test/compatibility/test.go
@@ -33,7 +33,7 @@
 
 // Parts of code were taken from prometheus repo: https://github.com/prometheus/prometheus/blob/master/promql/test.go
 
-package test
+package compatibility
 
 import (
 	"context"
@@ -48,10 +48,10 @@ import (
 	"time"
 
 	cparser "github.com/m3db/m3/src/cmd/services/m3comparator/main/parser"
+	"github.com/m3db/m3/src/query/api/v1/handler/prometheus"
 
 	"github.com/pkg/errors"
 	"github.com/prometheus/common/model"
-
 	"github.com/prometheus/prometheus/pkg/labels"
 	"github.com/prometheus/prometheus/promql"
 	"github.com/prometheus/prometheus/promql/parser"
@@ -391,15 +391,13 @@ func (ev *evalCmd) expect(pos int, m labels.Labels, vals ...parser.SequenceValue
 	ev.expected[h] = entry{pos: pos, vals: vals}
 }
 
-const sep = '\xff'
-
 // Hash returns a hash value for the label set.
-func hash(ls model.Metric) uint64 {
+func hash(ls prometheus.Tags) uint64 {
 	lbs := make(labels.Labels, 0, len(ls))
 	for k, v := range ls {
 		lbs = append(lbs, labels.Label{
-			Name:  string(k),
-			Value: string(v),
+			Name:  k,
+			Value: v,
 		})
 	}
 
@@ -410,34 +408,27 @@ func hash(ls model.Metric) uint64 {
 	return lbs.Hash()
 }
 
-// QueryResponse defines a structure for expected response.
-type QueryResponse struct {
-	Status string
-	Data   struct {
-		ResultType string
-		Result     model.Samples
-	}
-}
-
 // compareResult compares the result value with the defined expectation.
 func (ev *evalCmd) compareResult(j []byte) error {
-	var result QueryResponse
-	err := json.Unmarshal(j, &result)
+	var response prometheus.Response
+	err := json.Unmarshal(j, &response)
 	if err != nil {
 		return err
 	}
 
-	if result.Status != "success" {
-		return fmt.Errorf("unsuccess status received: %s", result.Status)
+	if response.Status != "success" {
+		return fmt.Errorf("unsuccess status received: %s", response.Status)
 	}
 
-	switch result.Data.ResultType {
-	case "matrix":
+	result := response.Data.Result
+
+	switch result := result.(type) {
+	case *prometheus.MatrixResult:
 		return errors.New("received range result on instant evaluation")
 
-	case "vector":
+	case *prometheus.VectorResult:
 		seen := map[uint64]bool{}
-		for pos, v := range result.Data.Result {
+		for pos, v := range result.Result {
 			fp := hash(v.Metric)
 			if _, ok := ev.metrics[fp]; !ok {
 				return errors.Errorf("unexpected metric %s in result", v.Metric)
@@ -447,31 +438,43 @@ func (ev *evalCmd) compareResult(j []byte) error {
 			if ev.ordered && exp.pos != pos+1 {
 				return errors.Errorf("expected metric %s with %v at position %d but was at %d", v.Metric, exp.vals, exp.pos, pos+1)
 			}
-			if !almostEqual(exp.vals[0].Value, float64(v.Value)) {
-				return errors.Errorf("expected %v for %s but got %v", exp.vals[0].Value, v.Metric, v.Value)
+			val, err := parseNumber(fmt.Sprint(v.Value[1]))
+			if err != nil {
+				return err
+			}
+			if !almostEqual(exp.vals[0].Value, val) {
+				return errors.Errorf("expected %v for %s but got %v", exp.vals[0].Value, v.Metric, val)
 			}
 			seen[fp] = true
 		}
 
 		for fp, expVals := range ev.expected {
 			if !seen[fp] {
-				fmt.Println("vector result", len(result.Data.Result), ev.expr)
-				for _, ss := range result.Data.Result {
+				fmt.Println("vector result", len(result.Result), ev.expr)
+				for _, ss := range result.Result {
 					fmt.Println("    ", ss.Metric, ss.Value)
 				}
 				return errors.Errorf("expected metric %s with %v not found", ev.metrics[fp], expVals)
 			}
 		}
 
-	// TODO: Untested code. Uncomment when m3query will support scalars and fix it if needed.
-	//case "scalar":
-	// if !almostEqual(ev.expected[0].vals[0].Value, val.V) {
-	// 	return errors.Errorf("expected Scalar %v but got %v", val.V, ev.expected[0].vals[0].Value)
-	// }
+	case *prometheus.ScalarResult:
+		v, err := parseNumber(fmt.Sprint(result.Result[1]))
+		if err != nil {
+			return err
+		}
+		if len(ev.expected) == 0 || len(ev.expected[0].vals) == 0 {
+			return errors.Errorf("expected no Scalar value but got %v", v)
+		}
+		expected := ev.expected[0].vals[0].Value
+		if !almostEqual(expected, v) {
+			return errors.Errorf("expected Scalar %v but got %v", expected, v)
+		}
 
 	default:
 		panic(errors.Errorf("promql.Test.compareResult: unexpected result type %T", result))
 	}
+
 	return nil
 }
 
@@ -535,8 +538,7 @@ func (t *Test) exec(tc testCommand) error {
 
 // clear the current test storage of all inserted samples.
 func (t *Test) clear() error {
-	t.m3comparator.clear()
-	return nil
+	return t.m3comparator.clear()
 }
 
 // Close closes resources associated with the Test.
@@ -560,7 +562,7 @@ func almostEqual(a, b float64) bool {
 	diff := math.Abs(a - b)
 
 	if a == 0 || b == 0 || diff < minNormal {
-		return diff < epsilon*minNormal
+		return diff < epsilon
 	}
 	return diff/(math.Abs(a)+math.Abs(b)) < epsilon
 }
diff --git a/src/query/test/testdata/LICENSE.txt b/src/query/test/compatibility/testdata/LICENSE.txt
similarity index 100%
rename from src/query/test/testdata/LICENSE.txt
rename to src/query/test/compatibility/testdata/LICENSE.txt
diff --git a/src/query/test/testdata/aggregators.test b/src/query/test/compatibility/testdata/aggregators.test
similarity index 90%
rename from src/query/test/testdata/aggregators.test
rename to src/query/test/compatibility/testdata/aggregators.test
index 8505cf18f1..716395e8b7 100644
--- a/src/query/test/testdata/aggregators.test
+++ b/src/query/test/compatibility/testdata/aggregators.test
@@ -36,10 +36,10 @@ eval instant at 50m count by (group) (http_requests{job="api-server"})
   {group="canary"} 2
   {group="production"} 2
 
-# FAILING issue #9. Simple without.
-#eval instant at 50m sum without (instance) (http_requests{job="api-server"})
-#  {group="canary",job="api-server"} 700
-#  {group="production",job="api-server"} 300
+# Simple without.
+eval instant at 50m sum without (instance) (http_requests{job="api-server"})
+  {group="canary",job="api-server"} 700
+  {group="production",job="api-server"} 300
 
 # Empty by.
 eval instant at 50m sum by () (http_requests{job="api-server"})
@@ -49,17 +49,17 @@ eval instant at 50m sum by () (http_requests{job="api-server"})
 eval instant at 50m sum(http_requests{job="api-server"})
   {} 1000
 
-# FAILING issue #9. Empty without.
-#eval instant at 50m sum without () (http_requests{job="api-server",group="production"})
-#  {group="production",job="api-server",instance="0"} 100
-#  {group="production",job="api-server",instance="1"} 200
+# Empty without.
+eval instant at 50m sum without () (http_requests{job="api-server",group="production"})
+  {group="production",job="api-server",instance="0"} 100
+  {group="production",job="api-server",instance="1"} 200
 
-# FAILING issue #9. Without with mismatched and missing labels. Do not do this.
-#eval instant at 50m sum without (instance) (http_requests{job="api-server"} or foo)
-#  {group="canary",job="api-server"} 700
-#  {group="production",job="api-server"} 300
-#  {region="europe",job="api-server"} 900
-#  {job="api-server"} 1000
+# Without with mismatched and missing labels. Do not do this.
+eval instant at 50m sum without (instance) (http_requests{job="api-server"} or foo)
+  {group="canary",job="api-server"} 700
+  {group="production",job="api-server"} 300
+  {region="europe",job="api-server"} 900
+  {job="api-server"} 1000
 
 # Lower-cased aggregation operators should work too.
 eval instant at 50m sum(http_requests) by (job) + min(http_requests) by (job) + max(http_requests) by (job) + avg(http_requests) by (job)
@@ -101,9 +101,11 @@ load 5m
   http_requests{job="api-server", instance="1", group="production"} 0+1.33x10
   http_requests{job="api-server", instance="0", group="canary"} 0+1.33x10
 
+# FAILING issue #10. (it is almost zero)
 #eval instant at 50m stddev(http_requests)
 #  {} 0.0
 
+# FAILING issue #11. (it is almost zero)
 #eval instant at 50m stdvar(http_requests)
 #  {} 0.0
 
@@ -115,10 +117,9 @@ load 5m
   label_grouping_test{a="aa", b="bb"} 0+10x10
   label_grouping_test{a="a", b="abb"} 0+20x10
 
-# FAILING
-#eval instant at 50m sum(label_grouping_test) by (a, b)
-#  {a="a", b="abb"} 200
-#  {a="aa", b="bb"} 100
+eval instant at 50m sum(label_grouping_test) by (a, b)
+  {a="a", b="abb"} 200
+  {a="aa", b="bb"} 100
 
 
 
@@ -184,9 +185,9 @@ load 5m
 #	http_requests{group="canary", instance="0", job="app-server"} 700
 #	http_requests{group="canary", instance="1", job="app-server"} 800
 
-#eval instant at 50m topk by (group) (1, http_requests)
-#  http_requests{group="production", instance="1", job="app-server"} 600
-#  http_requests{group="canary", instance="1", job="app-server"} 800
+eval instant at 50m topk by (group) (1, http_requests)
+  http_requests{group="production", instance="1", job="app-server"} 600
+  http_requests{group="canary", instance="1", job="app-server"} 800
 
 #eval instant at 50m bottomk by (group) (2, http_requests)
 #  http_requests{group="canary", instance="0", job="api-server"} 300
@@ -239,7 +240,7 @@ load 5m
 	version{job="app-server", instance="0", group="canary"}		7
 	version{job="app-server", instance="1", group="canary"}		7
 
-# FAILING issue #14
+# FAILING issue #14 (count_values)
 #eval instant at 5m count_values("version", version)
 #	{version="6"} 5
 #	{version="7"} 2
@@ -285,7 +286,7 @@ load 10s
 	data{test="uneven samples",point="c"} 4
 	foo .8
 
-# FAILING issue #8
+# FAILING issue #8 (quantile)
 #eval instant at 1m quantile without(point)(0.8, data)
 #	{test="two samples"} 0.8
 #	{test="three samples"} 1.6
diff --git a/src/query/test/testdata/functions.test b/src/query/test/compatibility/testdata/functions.test
similarity index 86%
rename from src/query/test/testdata/functions.test
rename to src/query/test/compatibility/testdata/functions.test
index 9f7f769cb1..bef714ed38 100644
--- a/src/query/test/testdata/functions.test
+++ b/src/query/test/compatibility/testdata/functions.test
@@ -26,7 +26,7 @@ eval instant at 50m resets(http_requests[50m])
 	{path="/bar"} 1
 	{path="/biz"} 0
 
-# FAILING. eval instant at 50m resets(nonexistent_metric[50m])
+eval instant at 50m resets(nonexistent_metric[50m])
 
 # Tests for changes().
 # FAILING issue #17
@@ -55,7 +55,7 @@ eval instant at 50m changes(http_requests[30m])
 #	{path="/bar"} 9
 #	{path="/biz"} 1
 
-# FAILING. eval instant at 50m changes(nonexistent_metric[50m])
+eval instant at 50m changes(nonexistent_metric[50m])
 
 clear
 
@@ -141,12 +141,10 @@ load 5m
 	testcounter_reset_middle	0+10x4 0+10x5
 	http_requests{job="app-server", instance="1", group="canary"}		0+80x10
 
-# FAILING TODO: should not fail, need to check
 # deriv should return the same as rate in simple cases.
 eval instant at 50m rate(http_requests{group="canary", instance="1", job="app-server"}[50m])
 	{group="canary", instance="1", job="app-server"} 0.26666666666666666
 
-# FAILING TODO: should not fail, need to check
 eval instant at 50m deriv(http_requests{group="canary", instance="1", job="app-server"}[50m])
 	{group="canary", instance="1", job="app-server"} 0.26666666666666666
 
@@ -241,16 +239,16 @@ clear
 load 10s
   metric 1 1
 
-# FAILING. eval instant at 0s timestamp(metric)
+# FAILING issue #23. eval instant at 0s timestamp(metric)
 #  {} 0
 
-# FAILING. eval instant at 5s timestamp(metric)
+# FAILING issue #23. eval instant at 5s timestamp(metric)
 #  {} 0
 
-# FAILING. eval instant at 10s timestamp(metric)
+# FAILING issue #23. eval instant at 10s timestamp(metric)
 #  {} 10
 
-# FAILING. eval instant at 10s timestamp(((metric)))
+# FAILING issue #23. eval instant at 10s timestamp(((metric)))
 #  {} 10
 
 # Tests for label_join.
@@ -344,7 +342,8 @@ load 5m
 	http_requests{job="app-server", instance="0", group="canary"}		0+70x10
 	http_requests{job="app-server", instance="1", group="canary"}		0+80x10
 
-# FAILING. eval_ordered instant at 50m sort(http_requests)
+# FAILING issue #28:
+#eval_ordered instant at 50m sort(http_requests)
 #	http_requests{group="production", instance="0", job="api-server"} 100
 #	http_requests{group="production", instance="1", job="api-server"} 200
 #	http_requests{group="canary", instance="0", job="api-server"} 300
@@ -415,7 +414,7 @@ eval instant at 1m stdvar_over_time(metric[1m])
 eval instant at 1m stddev_over_time(metric[1m])
   {} 3.249615
 
-# FAILING. eval instant at 1m stddev_over_time((metric[1m]))
+# FAILING issue #19. eval instant at 1m stddev_over_time((metric[1m]))
 #  {} 3.249615
 
 # Tests for stddev_over_time and stdvar_over_time #4927.
@@ -472,14 +471,14 @@ eval instant at 1m quantile_over_time(2, data[1m])
 	{test="three samples"} +Inf
 	{test="uneven samples"} +Inf
 
-# FAILING. eval instant at 1m (quantile_over_time(2, (data[1m])))
+# FAILING issue #19. eval instant at 1m (quantile_over_time(2, (data[1m])))
 #	{test="two samples"} +Inf
 #	{test="three samples"} +Inf
 #	{test="uneven samples"} +Inf
 
 clear
 
-# FAILING. Test time-related functions.
+# FAILING issue #30. Test time-related functions.
 #eval instant at 0m year()
 #  {} 1970
 
@@ -579,23 +578,23 @@ load 10s
 
 clear
 
-# FAILING. Testdata for absent_over_time()
+# FAILING issue #6. Testdata for absent_over_time()
 #eval instant at 1m absent_over_time(http_requests[5m])
 #    {} 1
 
-# FAILING. eval instant at 1m absent_over_time(http_requests{handler="/foo"}[5m])
+# FAILING issue #6. eval instant at 1m absent_over_time(http_requests{handler="/foo"}[5m])
 #    {handler="/foo"} 1
 
-# FAILING. eval instant at 1m absent_over_time(http_requests{handler!="/foo"}[5m])
+# FAILING issue #6. eval instant at 1m absent_over_time(http_requests{handler!="/foo"}[5m])
 #    {} 1
 
-# FAILING. eval instant at 1m absent_over_time(http_requests{handler="/foo", handler="/bar", handler="/foobar"}[5m])
+# FAILING issue #6. eval instant at 1m absent_over_time(http_requests{handler="/foo", handler="/bar", handler="/foobar"}[5m])
 #    {} 1
 
-# FAILING. eval instant at 1m absent_over_time(rate(nonexistant[5m])[5m:])
+# FAILING issue #6. eval instant at 1m absent_over_time(rate(nonexistant[5m])[5m:])
 #    {} 1
 
-# FAILING. eval instant at 1m absent_over_time(http_requests{handler="/foo", handler="/bar", instance="127.0.0.1"}[5m])
+# FAILING issue #6. eval instant at 1m absent_over_time(http_requests{handler="/foo", handler="/bar", instance="127.0.0.1"}[5m])
 #    {instance="127.0.0.1"} 1
 
 load 1m
@@ -605,40 +604,40 @@ load 1m
 	httpd_log_lines_total{instance="127.0.0.1",job="node"}	1
 	ssl_certificate_expiry_seconds{job="ingress"} NaN NaN NaN NaN NaN
 
-# FAILING. eval instant at 5m absent_over_time(http_requests[5m])
+# FAILING issue #6. eval instant at 5m absent_over_time(http_requests[5m])
 
-# FAILING. eval instant at 5m absent_over_time(rate(http_requests[5m])[5m:1m])
+# FAILING issue #6. eval instant at 5m absent_over_time(rate(http_requests[5m])[5m:1m])
 
-# FAILING. eval instant at 0m absent_over_time(httpd_log_lines_total[30s])
+# FAILING issue #6. eval instant at 0m absent_over_time(httpd_log_lines_total[30s])
 
-# FAILING. eval instant at 1m absent_over_time(httpd_log_lines_total[30s])
+# FAILING issue #6. eval instant at 1m absent_over_time(httpd_log_lines_total[30s])
 #    {} 1
 
-# FAILING. eval instant at 15m absent_over_time(http_requests[5m])
+# FAILING issue #6. eval instant at 15m absent_over_time(http_requests[5m])
 
-# FAILING. eval instant at 16m absent_over_time(http_requests[5m])
+# FAILING issue #6. eval instant at 16m absent_over_time(http_requests[5m])
 #    {} 1
 
-# FAILING. eval instant at 16m absent_over_time(http_requests[6m])
+# FAILING issue #6. eval instant at 16m absent_over_time(http_requests[6m])
 
-# FAILING. eval instant at 16m absent_over_time(httpd_handshake_failures_total[1m])
+# FAILING issue #6. eval instant at 16m absent_over_time(httpd_handshake_failures_total[1m])
 
-# FAILING. eval instant at 16m absent_over_time({instance="127.0.0.1"}[5m])
+# FAILING issue #6. eval instant at 16m absent_over_time({instance="127.0.0.1"}[5m])
 
-# FAILING. eval instant at 16m absent_over_time({instance="127.0.0.1"}[5m])
+# FAILING issue #6. eval instant at 16m absent_over_time({instance="127.0.0.1"}[5m])
 
-# FAILING. eval instant at 21m absent_over_time({instance="127.0.0.1"}[5m])
-# FAILING.     {instance="127.0.0.1"} 1
+# FAILING issue #6. eval instant at 21m absent_over_time({instance="127.0.0.1"}[5m])
+# FAILING issue #6.     {instance="127.0.0.1"} 1
 
-# FAILING. eval instant at 21m absent_over_time({instance="127.0.0.1"}[20m])
+# FAILING issue #6. eval instant at 21m absent_over_time({instance="127.0.0.1"}[20m])
 
-# FAILING. eval instant at 21m absent_over_time({job="grok"}[20m])
-# FAILING.     {job="grok"} 1
+# FAILING issue #6. eval instant at 21m absent_over_time({job="grok"}[20m])
+# FAILING issue #6.     {job="grok"} 1
 
-# FAILING. eval instant at 30m absent_over_time({instance="127.0.0.1"}[5m:5s])
-# FAILING.     {} 1
+# FAILING issue #6. eval instant at 30m absent_over_time({instance="127.0.0.1"}[5m:5s])
+# FAILING issue #6.     {} 1
 
-# FAILING. eval instant at 5m absent_over_time({job="ingress"}[4m])
+# FAILING issue #6. eval instant at 5m absent_over_time({job="ingress"}[4m])
 
-# FAILING. eval instant at 10m absent_over_time({job="ingress"}[4m])
-# FAILING. 	{job="ingress"} 1
+# FAILING issue #6. eval instant at 10m absent_over_time({job="ingress"}[4m])
+# FAILING issue #6. 	{job="ingress"} 1
diff --git a/src/query/test/testdata/histograms.test b/src/query/test/compatibility/testdata/histograms.test
similarity index 97%
rename from src/query/test/testdata/histograms.test
rename to src/query/test/compatibility/testdata/histograms.test
index 88fe483e64..664fd6c4ed 100644
--- a/src/query/test/testdata/histograms.test
+++ b/src/query/test/compatibility/testdata/histograms.test
@@ -164,7 +164,8 @@ load 5m
     nonmonotonic_bucket{le="+Inf"}  0+9x10
 
 # Nonmonotonic buckets
-# FAILING issue #48. eval instant at 50m histogram_quantile(0.99, nonmonotonic_bucket)
+# TODO - enable after https://github.com/prometheus/prometheus/commit/7eaffa71802821087c8ee42dbda652453ec1f867 is released:
+# eval instant at 50m histogram_quantile(0.99, nonmonotonic_bucket)
 #    {} 0.989875
 
 # FAILING issue #48. Buckets with different representations of the same upper bound.
diff --git a/src/query/test/testdata/legacy.test b/src/query/test/compatibility/testdata/legacy.test
similarity index 78%
rename from src/query/test/testdata/legacy.test
rename to src/query/test/compatibility/testdata/legacy.test
index b60a352ed1..10a7143ae5 100644
--- a/src/query/test/testdata/legacy.test
+++ b/src/query/test/compatibility/testdata/legacy.test
@@ -229,69 +229,69 @@ eval instant at 50m sum_over_time(http_requests{group="production",job="api-serv
 # FAILING. eval instant at 50m time()
 #	3000
 
-# FAILING. eval instant at 50m {__name__=~".+"}
-#	http_requests{group="canary", instance="0", job="api-server"} 300
-#	http_requests{group="canary", instance="0", job="app-server"} 700
-#	http_requests{group="canary", instance="1", job="api-server"} 400
-#	http_requests{group="canary", instance="1", job="app-server"} 800
-#	http_requests{group="production", instance="0", job="api-server"} 100
-#	http_requests{group="production", instance="0", job="app-server"} 500
-#	http_requests{group="production", instance="1", job="api-server"} 200
-#	http_requests{group="production", instance="1", job="app-server"} 600
-#	testcounter_reset_end 0
-#	testcounter_reset_middle 50
-#	x{y="testvalue"} 100
-#	label_grouping_test{a="a", b="abb"} 200
-#	label_grouping_test{a="aa", b="bb"} 100
-#	vector_matching_a{l="x"} 10
-#	vector_matching_a{l="y"} 20
-#	vector_matching_b{l="x"} 40
-#	cpu_count{instance="1", type="smp"} 200
-#	cpu_count{instance="0", type="smp"} 100
-#	cpu_count{instance="0", type="numa"} 300
-
-
-# FAILING. eval instant at 50m {job=~".+-server", job!~"api-.+"}
-#	http_requests{group="canary", instance="0", job="app-server"} 700
-#	http_requests{group="canary", instance="1", job="app-server"} 800
-#	http_requests{group="production", instance="0", job="app-server"} 500
-#	http_requests{group="production", instance="1", job="app-server"} 600
-
-# FAILING. eval instant at 50m absent(nonexistent)
+eval instant at 50m {__name__=~".+"}
+	http_requests{group="canary", instance="0", job="api-server"} 300
+	http_requests{group="canary", instance="0", job="app-server"} 700
+	http_requests{group="canary", instance="1", job="api-server"} 400
+	http_requests{group="canary", instance="1", job="app-server"} 800
+	http_requests{group="production", instance="0", job="api-server"} 100
+	http_requests{group="production", instance="0", job="app-server"} 500
+	http_requests{group="production", instance="1", job="api-server"} 200
+	http_requests{group="production", instance="1", job="app-server"} 600
+	testcounter_reset_end 0
+	testcounter_reset_middle 50
+	x{y="testvalue"} 100
+	label_grouping_test{a="a", b="abb"} 200
+	label_grouping_test{a="aa", b="bb"} 100
+	vector_matching_a{l="x"} 10
+	vector_matching_a{l="y"} 20
+	vector_matching_b{l="x"} 40
+	cpu_count{instance="1", type="smp"} 200
+	cpu_count{instance="0", type="smp"} 100
+	cpu_count{instance="0", type="numa"} 300
+
+
+eval instant at 50m {job=~".+-server", job!~"api-.+"}
+	http_requests{group="canary", instance="0", job="app-server"} 700
+	http_requests{group="canary", instance="1", job="app-server"} 800
+	http_requests{group="production", instance="0", job="app-server"} 500
+	http_requests{group="production", instance="1", job="app-server"} 600
+
+# eval instant at 50m absent(nonexistent)
 #	{} 1
 
-# FAILING. eval instant at 50m absent(nonexistent{job="testjob", instance="testinstance", method=~".x"})
+# FAILING issue #52. eval instant at 50m absent(nonexistent{job="testjob", instance="testinstance", method=~".x"})
 #	{instance="testinstance", job="testjob"} 1
 
-# FAILING. eval instant at 50m absent(nonexistent{job="testjob",job="testjob2",foo="bar"})
+# FAILING issue #52. eval instant at 50m absent(nonexistent{job="testjob",job="testjob2",foo="bar"})
 #	{foo="bar"} 1
 
-# FAILING. eval instant at 50m absent(nonexistent{job="testjob",job="testjob2",job="three",foo="bar"})
+# FAILING issue #52. eval instant at 50m absent(nonexistent{job="testjob",job="testjob2",job="three",foo="bar"})
 #	{foo="bar"} 1
 
-# FAILING. eval instant at 50m absent(nonexistent{job="testjob",job=~"testjob2",foo="bar"})
+# FAILING issue #52. eval instant at 50m absent(nonexistent{job="testjob",job=~"testjob2",foo="bar"})
 #	{foo="bar"} 1
 
 eval instant at 50m absent(http_requests)
 
 eval instant at 50m absent(sum(http_requests))
 
-# FAILING. eval instant at 50m absent(sum(nonexistent{job="testjob", instance="testinstance"}))
+# FAILING issue #52. eval instant at 50m absent(sum(nonexistent{job="testjob", instance="testinstance"}))
 #	{} 1
 
-# FAILING. eval instant at 50m absent(max(nonexistant))
+# FAILING issue #52. eval instant at 50m absent(max(nonexistant))
 #	{} 1
 
-# FAILING. eval instant at 50m absent(nonexistant > 1)
+# FAILING issue #52. eval instant at 50m absent(nonexistant > 1)
 #	{} 1
 
-# FAILING. eval instant at 50m absent(a + b)
+# FAILING issue #52. eval instant at 50m absent(a + b)
 #	{} 1
 
-# FAILING. eval instant at 50m absent(a and b)
+# FAILING issue #52. eval instant at 50m absent(a and b)
 #	{} 1
 
-# FAILING. eval instant at 50m absent(rate(nonexistant[5m]))
+# FAILING issue #52. eval instant at 50m absent(rate(nonexistant[5m]))
 #	{} 1
 
 eval instant at 50m http_requests{group="production",job="api-server"} offset 5m
@@ -314,7 +314,7 @@ eval instant at 50m -1 * http_requests{group="canary", instance="0", job="api-se
 # Failing with keepNaN feature. eval instant at 50m 0 * http_requests{group="canary", instance="0", job="api-server"} % 0
 #	{group="canary", instance="0", job="api-server"} NaN
 
-# FAILING. eval instant at 50m exp(vector_matching_a)
+# FAILING issue #53. eval instant at 50m exp(vector_matching_a)
 #	{l="x"} 22026.465794806718
 #	{l="y"} 485165195.4097903
 
@@ -326,53 +326,54 @@ eval instant at 50m exp(vector_matching_a - 20)
 	{l="x"} 4.5399929762484854e-05
 	{l="y"} 1
 
-# FAILING. eval instant at 50m ln(vector_matching_a)
+# FAILING issue #53. eval instant at 50m ln(vector_matching_a)
 #	{l="x"} 2.302585092994046
 #	{l="y"} 2.995732273553991
 
-# FAILING. eval instant at 50m ln(vector_matching_a - 10)
-#	{l="y"} 2.302585092994046
-#	{l="x"} -Inf
+eval instant at 50m ln(vector_matching_a - 10)
+	{l="y"} 2.302585092994046
+	{l="x"} -Inf
 
-# FAILING. eval instant at 50m ln(vector_matching_a - 20)
+# FAILING with keepNaN feature. eval instant at 50m ln(vector_matching_a - 20)
 #	{l="y"} -Inf
 #	{l="x"} NaN
 
-# FAILING. eval instant at 50m exp(ln(vector_matching_a))
+# FAILING issue #53. eval instant at 50m exp(ln(vector_matching_a))
 #	{l="y"} 20
 #	{l="x"} 10
 
-# FAILING. eval instant at 50m sqrt(vector_matching_a)
+# FAILING issue #53. eval instant at 50m sqrt(vector_matching_a)
 #	{l="x"} 3.1622776601683795
 #	{l="y"} 4.47213595499958
 
-# FAILING. eval instant at 50m log2(vector_matching_a)
+# FAILING issue #53. eval instant at 50m log2(vector_matching_a)
 #	{l="x"} 3.3219280948873626
 #	{l="y"} 4.321928094887363
 
-# FAILING. eval instant at 50m log2(vector_matching_a - 10)
-#	{l="y"} 3.3219280948873626
-#	{l="x"} -Inf
+eval instant at 50m log2(vector_matching_a - 10)
+	{l="y"} 3.3219280948873626
+	{l="x"} -Inf
 
-# FAILING. eval instant at 50m log2(vector_matching_a - 20)
+# FAILING with keepNaN feature. eval instant at 50m log2(vector_matching_a - 20)
 #	{l="x"} NaN
 #	{l="y"} -Inf
 
-# FAILING. eval instant at 50m log10(vector_matching_a)
+# FAILING issue #53. eval instant at 50m log10(vector_matching_a)
 #	{l="x"} 1
 #	{l="y"} 1.301029995663981
 
-# FAILING. eval instant at 50m log10(vector_matching_a - 10)
-#	{l="y"} 1
-#	{l="x"} -Inf
+eval instant at 50m log10(vector_matching_a - 10)
+	{l="y"} 1
+	{l="x"} -Inf
 
-# FAILING. eval instant at 50m log10(vector_matching_a - 20)
+# FAILING with keepNaN feature. eval instant at 50m log10(vector_matching_a - 20)
 #	{l="x"} NaN
 #	{l="y"} -Inf
 
 
 # Matrix tests.
 clear
+
 load 1h
 	testmetric{aa="bb"} 1
 	testmetric{a="abb"} 2
diff --git a/src/query/test/compatibility/testdata/literals.test b/src/query/test/compatibility/testdata/literals.test
new file mode 100644
index 0000000000..346d792610
--- /dev/null
+++ b/src/query/test/compatibility/testdata/literals.test
@@ -0,0 +1,59 @@
+eval instant at 50m 12.34e6
+    12340000
+
+eval instant at 50m 12.34e+6
+	12340000
+
+eval instant at 50m 12.34e-6
+	0.00001234
+
+eval instant at 50m 1+1
+	2
+
+eval instant at 50m 1-1
+	0
+
+eval instant at 50m 1 - -1
+	2
+
+eval instant at 50m .2
+	0.2
+
+eval instant at 50m +0.2
+	0.2
+
+eval instant at 50m -0.2e-6
+	-0.0000002
+
+eval instant at 50m +Inf
+	+Inf
+
+eval instant at 50m inF
+	+Inf
+
+eval instant at 50m -inf
+	-Inf
+
+eval instant at 50m NaN
+	NaN
+
+eval instant at 50m nan
+	NaN
+
+eval instant at 50m 2.
+	2
+
+eval instant at 50m 1 / 0
+	+Inf
+
+eval instant at 50m ((1) / (0))
+	+Inf
+
+eval instant at 50m -1 / 0
+	-Inf
+
+eval instant at 50m 0 / 0
+	NaN
+
+eval instant at 50m 1 % 0
+	NaN
diff --git a/src/query/test/testdata/operators.test b/src/query/test/compatibility/testdata/operators.test
similarity index 94%
rename from src/query/test/testdata/operators.test
rename to src/query/test/compatibility/testdata/operators.test
index a0ecb8779e..a7c541af62 100644
--- a/src/query/test/testdata/operators.test
+++ b/src/query/test/compatibility/testdata/operators.test
@@ -22,7 +22,7 @@ eval instant at 50m 2 - SUM(http_requests) BY (job)
 	{job="api-server"} -998
 	{job="app-server"} -2598
 
-# FAILING. eval instant at 50m -http_requests{job="api-server",instance="0",group="production"}
+# FAILING issue #53. eval instant at 50m -http_requests{job="api-server",instance="0",group="production"}
 #  {job="api-server",instance="0",group="production"} -100
 
 eval instant at 50m +http_requests{job="api-server",instance="0",group="production"}
@@ -32,14 +32,14 @@ eval instant at 50m - - - SUM(http_requests) BY (job)
 	{job="api-server"} -1000
 	{job="app-server"} -2600
 
-# FAILING. eval instant at 50m - - - 1
-#  -1
+eval instant at 50m - - - 1
+  -1
 
 # FAILING. eval instant at 50m -2^---1*3
 #  -1.5
 
 # FAILING. eval instant at 50m 2/-2^---1*3+2
-#  -10 
+#  -10
 
 # FAILING. eval instant at 50m -10^3 * - SUM(http_requests) BY (job) ^ -1
 #	{job="api-server"} 1
@@ -115,7 +115,7 @@ eval instant at 50m rate(http_requests[25m]) * 25 * 60
   {group="production", instance="1", job="api-server"} 100
   {group="production", instance="1", job="app-server"} 300
 
-# FAILING issue #30. eval instant at 50m (rate((http_requests[25m])) * 25) * 60
+# FAILING issue #19. eval instant at 50m (rate((http_requests[25m])) * 25) * 60
 #  {group="canary", instance="0", job="api-server"} 150
 #  {group="canary", instance="0", job="app-server"} 350
 #  {group="canary", instance="1", job="api-server"} 200
@@ -126,7 +126,7 @@ eval instant at 50m rate(http_requests[25m]) * 25 * 60
 #  {group="production", instance="1", job="app-server"} 300
 
 
-# FAILING issue #23. eval instant at 50m http_requests{group="canary"} and http_requests{instance="0"}
+# FAILING order of rows differs. eval instant at 50m http_requests{group="canary"} and http_requests{instance="0"}
 #	http_requests{group="canary", instance="0", job="api-server"} 300
 #	http_requests{group="canary", instance="0", job="app-server"} 700
 
@@ -150,7 +150,7 @@ eval instant at 50m (http_requests{group="canary"} + 1) and ignoring(group, job)
 	{group="canary", instance="0", job="api-server"} 301
 	{group="canary", instance="0", job="app-server"} 701
 
-# FAILING issue #23. eval instant at 50m http_requests{group="canary"} or http_requests{group="production"}
+# FAILING order of rows differs. eval instant at 50m http_requests{group="canary"} or http_requests{group="production"}
 #	http_requests{group="canary", instance="0", job="api-server"} 300
 #	http_requests{group="canary", instance="0", job="app-server"} 700
 #	http_requests{group="canary", instance="1", job="api-server"} 400
@@ -239,20 +239,20 @@ eval instant at 50m (http_requests{group="canary"} + 1) and ignoring(group, job)
 
 
 # Comparisons.
-# FAILING issue #37. eval instant at 50m SUM(http_requests) BY (job) > 1000
-#	{job="app-server"} 2600
+eval instant at 50m SUM(http_requests) BY (job) > 1000
+	{job="app-server"} 2600
 
-# FAILING issue #37. eval instant at 50m 1000 < SUM(http_requests) BY (job)
+# FAILING (returns lhs instead of rhs). eval instant at 50m 1000 < SUM(http_requests) BY (job)
 #	{job="app-server"} 2600
 
-# FAILING issue #37. eval instant at 50m SUM(http_requests) BY (job) <= 1000
-#	{job="api-server"} 1000
+eval instant at 50m SUM(http_requests) BY (job) <= 1000
+	{job="api-server"} 1000
 
-# FAILING issue #37. eval instant at 50m SUM(http_requests) BY (job) != 1000
-#	{job="app-server"} 2600
+eval instant at 50m SUM(http_requests) BY (job) != 1000
+	{job="app-server"} 2600
 
-# FAILING issue #37. eval instant at 50m SUM(http_requests) BY (job) == 1000
-#	{job="api-server"} 1000
+eval instant at 50m SUM(http_requests) BY (job) == 1000
+	{job="api-server"} 1000
 
 eval instant at 50m SUM(http_requests) BY (job) == bool 1000
 	{job="api-server"} 1
@@ -266,11 +266,11 @@ eval instant at 50m SUM(http_requests) BY (job) != bool SUM(http_requests) BY (j
 	{job="api-server"} 0
 	{job="app-server"} 0
 
-# FAILING issue #31. eval instant at 50m 0 == bool 1
-#	0
+eval instant at 50m 0 == bool 1
+	0
 
-# FAILING issue #31. eval instant at 50m 1 == bool 1
-#	1
+eval instant at 50m 1 == bool 1
+	1
 
 eval instant at 50m http_requests{job="api-server", instance="0", group="production"} == bool 100
 	{job="api-server", instance="0", group="production"} 1
@@ -420,7 +420,7 @@ load 5m
   metricB{baz="meh"} 4
 
 # On with no labels, for metrics with no common labels.
-# FAILING. eval instant at 5m random + on() metricA
+# FAILING issue #36. eval instant at 5m random + on() metricA
 #  {} 5
 
 # Ignoring with no labels is the same as no ignoring.
diff --git a/src/query/test/testdata/selectors.test b/src/query/test/compatibility/testdata/selectors.test
similarity index 100%
rename from src/query/test/testdata/selectors.test
rename to src/query/test/compatibility/testdata/selectors.test
diff --git a/src/query/test/testdata/staleness.test b/src/query/test/compatibility/testdata/staleness.test
similarity index 100%
rename from src/query/test/testdata/staleness.test
rename to src/query/test/compatibility/testdata/staleness.test
diff --git a/src/query/test/testdata/subquery.test b/src/query/test/compatibility/testdata/subquery.test
similarity index 100%
rename from src/query/test/testdata/subquery.test
rename to src/query/test/compatibility/testdata/subquery.test
diff --git a/src/query/test/testdata/literals.test b/src/query/test/testdata/literals.test
deleted file mode 100644
index 344e2a808a..0000000000
--- a/src/query/test/testdata/literals.test
+++ /dev/null
@@ -1,61 +0,0 @@
-# FAILING issue #46 (we return vector resultType in place of scalar).
-
-# FAILING. eval instant at 50m 12.34e6
-#	12340000
-
-# FAILING. eval instant at 50m 12.34e+6
-#	12340000
-
-# FAILING. eval instant at 50m 12.34e-6
-#	0.00001234
-
-# FAILING. eval instant at 50m 1+1
-#	2
-
-# FAILING. eval instant at 50m 1-1
-#	0
-
-# FAILING. eval instant at 50m 1 - -1
-#	2
-
-# FAILING. eval instant at 50m .2
-#	0.2
-
-# FAILING. eval instant at 50m +0.2
-#	0.2
-
-# FAILING. eval instant at 50m -0.2e-6
-#	-0.0000002
-
-# FAILING. eval instant at 50m +Inf
-#	+Inf
-
-# FAILING. eval instant at 50m inF
-#	+Inf
-
-# FAILING. eval instant at 50m -inf
-#	-Inf
-
-# FAILING. eval instant at 50m NaN
-#	NaN
-
-# FAILING. eval instant at 50m nan
-#	NaN
-
-# FAILING. eval instant at 50m 2.
-#	2
-
-# FAILING. eval instant at 50m 1 / 0
-#	+Inf
-
-# FAILING. eval instant at 50m ((1) / (0))
-#	+Inf
-
-# FAILING. eval instant at 50m -1 / 0
-#	-Inf
-
-# FAILING. eval instant at 50m 0 / 0
-#	NaN
-
-# FAILING. eval instant at 50m 1 % 0
-#	NaN